2. Training and Detection.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "QUANWN3rpfC9"
   },
   "source": [
    "# 0. Setup Paths"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "146BB11JpfDA"
   },
   "outputs": [],
   "source": [
    "import os"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "id": "42hJEdo_pfDB"
   },
   "outputs": [],
   "source": [
    "CUSTOM_MODEL_NAME = 'my_ssd_mobnet' \n",
    "PRETRAINED_MODEL_NAME = 'ssd_mobilenet_v2_fpnlite_320x320_coco17_tpu-8'\n",
    "PRETRAINED_MODEL_URL = 'http://download.tensorflow.org/models/object_detection/tf2/20200711/ssd_mobilenet_v2_fpnlite_320x320_coco17_tpu-8.tar.gz'\n",
    "TF_RECORD_SCRIPT_NAME = 'generate_tfrecord.py'\n",
    "LABEL_MAP_NAME = 'label_map.pbtxt'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "id": "hbPhYVy_pfDB"
   },
   "outputs": [],
   "source": [
    "paths = {\n",
    "    'WORKSPACE_PATH': os.path.join('Tensorflow', 'workspace'),\n",
    "    'SCRIPTS_PATH': os.path.join('Tensorflow','scripts'),\n",
    "    'APIMODEL_PATH': os.path.join('Tensorflow','models'),\n",
    "    'ANNOTATION_PATH': os.path.join('Tensorflow', 'workspace','annotations'),\n",
    "    'IMAGE_PATH': os.path.join('Tensorflow', 'workspace','images'),\n",
    "    'MODEL_PATH': os.path.join('Tensorflow', 'workspace','models'),\n",
    "    'PRETRAINED_MODEL_PATH': os.path.join('Tensorflow', 'workspace','pre-trained-models'),\n",
    "    'CHECKPOINT_PATH': os.path.join('Tensorflow', 'workspace','models',CUSTOM_MODEL_NAME), \n",
    "    'OUTPUT_PATH': os.path.join('Tensorflow', 'workspace','models',CUSTOM_MODEL_NAME, 'export'), \n",
    "    'TFJS_PATH':os.path.join('Tensorflow', 'workspace','models',CUSTOM_MODEL_NAME, 'tfjsexport'), \n",
    "    'TFLITE_PATH':os.path.join('Tensorflow', 'workspace','models',CUSTOM_MODEL_NAME, 'tfliteexport'), \n",
    "    'PROTOC_PATH':os.path.join('Tensorflow','protoc')\n",
    " }"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "id": "LwhWZMI0pfDC"
   },
   "outputs": [],
   "source": [
    "files = {\n",
    "    'PIPELINE_CONFIG':os.path.join('Tensorflow', 'workspace','models', CUSTOM_MODEL_NAME, 'pipeline.config'),\n",
    "    'TF_RECORD_SCRIPT': os.path.join(paths['SCRIPTS_PATH'], TF_RECORD_SCRIPT_NAME), \n",
    "    'LABELMAP': os.path.join(paths['ANNOTATION_PATH'], LABEL_MAP_NAME)\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "id": "HR-TfDGrpfDC"
   },
   "outputs": [],
   "source": [
    "for path in paths.values():\n",
    "    if not os.path.exists(path):\n",
    "        if os.name == 'posix':\n",
    "            !mkdir -p {path}\n",
    "        if os.name == 'nt':\n",
    "            !mkdir {path}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "OLU-rs_ipfDE"
   },
   "source": [
    "# 1. Download TF Models Pretrained Models from Tensorflow Model Zoo and Install TFOD"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# https://www.tensorflow.org/install/source_windows"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "id": "K-Cmz2edpfDE",
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: wget in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (3.2)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\absl_py-2.0.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\apache_beam-2.53.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\avro_python3-1.10.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\contextlib2-21.6.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\cython-3.0.8-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\dm_tree-0.1.8-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\h5py-3.10.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lvis-0.5.3-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lxml-5.1.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\namex-0.0.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\object_detection-0.1-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pandas-2.2.0rc0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pillow-10.2.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\portalocker-2.8.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pycocotools-2.0.7-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pyparsing-2.4.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\regex-2023.12.25-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\rich-13.7.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\sacrebleu-2.2.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\scipy-1.12.0rc2-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tabulate-0.9.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io_gcs_filesystem-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_models_official-2.15.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_slim-1.1.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n"
     ]
    }
   ],
   "source": [
    "if os.name=='nt':\n",
    "    !pip install wget\n",
    "    import wget"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "iA1DIq5OpfDE"
   },
   "outputs": [],
   "source": [
    "if not os.path.exists(os.path.join(paths['APIMODEL_PATH'], 'research', 'object_detection')):\n",
    "    !git clone https://github.com/tensorflow/models {paths['APIMODEL_PATH']}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "rJjMHbnDs3Tv"
   },
   "outputs": [],
   "source": [
    "# Install Tensorflow Object Detection \n",
    "if os.name=='posix':  \n",
    "    !apt-get install protobuf-compiler\n",
    "    !cd Tensorflow/models/research && protoc object_detection/protos/*.proto --python_out=. && cp object_detection/packages/tf2/setup.py . && python -m pip install . \n",
    "    \n",
    "if os.name=='nt':\n",
    "    url=\"https://github.com/protocolbuffers/protobuf/releases/download/v3.15.6/protoc-3.15.6-win64.zip\"\n",
    "    wget.download(url)\n",
    "    !move protoc-3.15.6-win64.zip {paths['PROTOC_PATH']}\n",
    "    !cd {paths['PROTOC_PATH']} && tar -xf protoc-3.15.6-win64.zip\n",
    "    os.environ['PATH'] += os.pathsep + os.path.abspath(os.path.join(paths['PROTOC_PATH'], 'bin'))   \n",
    "    !cd Tensorflow/models/research && protoc object_detection/protos/*.proto --python_out=. && copy object_detection\\\\packages\\\\tf2\\\\setup.py setup.py && python setup.py build && python setup.py install\n",
    "    !cd Tensorflow/models/research/slim && pip install -e . "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Package                      Version         Editable project location\n",
      "---------------------------- --------------- -------------------------------------------------------------\n",
      "absl-py                      2.0.0\n",
      "absl-py                      2.0.0\n",
      "anyio                        4.2.0\n",
      "apache-beam                  2.53.0\n",
      "apache-beam                  2.53.0\n",
      "argon2-cffi                  23.1.0\n",
      "argon2-cffi-bindings         21.2.0\n",
      "arrow                        1.3.0\n",
      "asttokens                    2.4.1\n",
      "astunparse                   1.6.3\n",
      "async-lru                    2.0.4\n",
      "attrs                        23.2.0\n",
      "avro-python3                 1.10.2\n",
      "avro-python3                 1.10.2\n",
      "avro-python3                 1.10.2\n",
      "Babel                        2.14.0\n",
      "beautifulsoup4               4.12.2\n",
      "bleach                       6.1.0\n",
      "cachetools                   5.3.2\n",
      "certifi                      2023.11.17\n",
      "cffi                         1.16.0\n",
      "charset-normalizer           3.3.2\n",
      "colorama                     0.4.6\n",
      "comm                         0.2.1\n",
      "contextlib2                  21.6.0\n",
      "contextlib2                  21.6.0\n",
      "contourpy                    1.2.0\n",
      "cycler                       0.12.1\n",
      "Cython                       3.0.8\n",
      "Cython                       3.0.8\n",
      "debugpy                      1.8.0\n",
      "decorator                    5.1.1\n",
      "defusedxml                   0.7.1\n",
      "dm-tree                      0.1.8\n",
      "dm-tree                      0.1.8\n",
      "easyocr                      1.7.1\n",
      "executing                    2.0.1\n",
      "fastjsonschema               2.19.1\n",
      "filelock                     3.9.0\n",
      "flatbuffers                  23.5.26\n",
      "fonttools                    4.47.2\n",
      "fqdn                         1.5.1\n",
      "fsspec                       2023.4.0\n",
      "gast                         0.4.0\n",
      "gin-config                   0.5.0\n",
      "google-auth                  2.26.2\n",
      "google-auth-oauthlib         1.0.0\n",
      "google-pasta                 0.2.0\n",
      "grpcio                       1.60.0\n",
      "h5py                         3.10.0\n",
      "h5py                         3.10.0\n",
      "idna                         3.6\n",
      "imageio                      2.33.1\n",
      "ipykernel                    6.28.0\n",
      "ipython                      8.20.0\n",
      "ipywidgets                   8.1.1\n",
      "isoduration                  20.11.0\n",
      "jedi                         0.19.1\n",
      "Jinja2                       3.1.3\n",
      "json5                        0.9.14\n",
      "jsonpointer                  2.4\n",
      "jsonschema                   4.20.0\n",
      "jsonschema-specifications    2023.12.1\n",
      "jupyter                      1.0.0\n",
      "jupyter_client               8.6.0\n",
      "jupyter-console              6.6.3\n",
      "jupyter_core                 5.7.1\n",
      "jupyter-events               0.9.0\n",
      "jupyter-lsp                  2.2.1\n",
      "jupyter_server               2.12.4\n",
      "jupyter_server_terminals     0.5.1\n",
      "jupyterlab                   4.0.10\n",
      "jupyterlab_pygments          0.3.0\n",
      "jupyterlab_server            2.25.2\n",
      "jupyterlab-widgets           3.0.9\n",
      "keras                        2.13.1\n",
      "kiwisolver                   1.4.5\n",
      "lazy_loader                  0.3\n",
      "libclang                     16.0.6\n",
      "lvis                         0.5.3\n",
      "lvis                         0.5.3\n",
      "lxml                         5.1.0\n",
      "lxml                         5.1.0\n",
      "Markdown                     3.5.2\n",
      "MarkupSafe                   2.1.3\n",
      "matplotlib-inline            0.1.6\n",
      "mistune                      3.0.2\n",
      "ml-dtypes                    0.2.0\n",
      "mpmath                       1.3.0\n",
      "namex                        0.0.7\n",
      "namex                        0.0.7\n",
      "nbclient                     0.9.0\n",
      "nbconvert                    7.14.1\n",
      "nbformat                     5.9.2\n",
      "nest-asyncio                 1.5.8\n",
      "networkx                     3.0\n",
      "ninja                        1.11.1.1\n",
      "notebook                     7.0.6\n",
      "notebook_shim                0.2.3\n",
      "numpy                        1.24.0\n",
      "oauthlib                     3.2.2\n",
      "object-detection             0.1\n",
      "object-detection             0.1\n",
      "opencv-python                4.9.0.80\n",
      "opt-einsum                   3.3.0\n",
      "overrides                    7.4.0\n",
      "packaging                    23.2\n",
      "pandas                       2.2.0rc0\n",
      "pandas                       2.2.0rc0\n",
      "pandocfilters                1.5.0\n",
      "parso                        0.8.3\n",
      "pillow                       10.2.0\n",
      "pillow                       10.2.0\n",
      "pip                          23.3.2\n",
      "platformdirs                 4.1.0\n",
      "portalocker                  2.8.2\n",
      "portalocker                  2.8.2\n",
      "prometheus-client            0.19.0\n",
      "prompt-toolkit               3.0.43\n",
      "protobuf                     4.25.2\n",
      "psutil                       5.9.7\n",
      "pure-eval                    0.2.2\n",
      "pyasn1                       0.5.1\n",
      "pyasn1-modules               0.3.0\n",
      "pyclipper                    1.3.0.post5\n",
      "pycocotools                  2.0.7\n",
      "pycocotools                  2.0.7\n",
      "pycparser                    2.21\n",
      "Pygments                     2.17.2\n",
      "pyparsing                    2.4.7\n",
      "pyparsing                    2.4.7\n",
      "python-bidi                  0.4.2\n",
      "python-dateutil              2.8.2\n",
      "python-json-logger           2.0.7\n",
      "pytz                         2023.3.post1\n",
      "pywin32                      306\n",
      "pywinpty                     2.0.12\n",
      "PyYAML                       6.0.1\n",
      "pyzmq                        25.1.2\n",
      "qtconsole                    5.5.1\n",
      "QtPy                         2.4.1\n",
      "referencing                  0.32.1\n",
      "regex                        2023.12.25\n",
      "regex                        2023.12.25\n",
      "requests                     2.31.0\n",
      "requests-oauthlib            1.3.1\n",
      "rfc3339-validator            0.1.4\n",
      "rfc3986-validator            0.1.1\n",
      "rich                         13.7.0\n",
      "rich                         13.7.0\n",
      "rpds-py                      0.17.1\n",
      "rsa                          4.9\n",
      "sacrebleu                    2.2.0\n",
      "sacrebleu                    2.2.0\n",
      "scikit-image                 0.22.0\n",
      "SciPy                        1.12.0rc2\n",
      "SciPy                        1.12.0rc2\n",
      "Send2Trash                   1.8.2\n",
      "setuptools                   65.5.0\n",
      "shapely                      2.0.2\n",
      "six                          1.16.0\n",
      "slim                         0.1             C:\\Users\\Hasitha\\Desktop\\ANPR\\Tensorflow\\models\\research\\slim\n",
      "sniffio                      1.3.0\n",
      "soupsieve                    2.5\n",
      "stack-data                   0.6.3\n",
      "sympy                        1.12\n",
      "tabulate                     0.9.0\n",
      "tabulate                     0.9.0\n",
      "tensorboard                  2.13.0\n",
      "tensorboard-data-server      0.7.2\n",
      "tensorflow                   2.13.0\n",
      "tensorflow-estimator         2.13.0\n",
      "tensorflow-intel             2.13.0\n",
      "tensorflow-io                0.31.0\n",
      "tensorflow-io                0.31.0\n",
      "tensorflow-io-gcs-filesystem 0.31.0\n",
      "tensorflow-io-gcs-filesystem 0.31.0\n",
      "termcolor                    2.4.0\n",
      "terminado                    0.18.0\n",
      "tf-models-official           2.15.0\n",
      "tf-models-official           2.15.0\n",
      "tf-slim                      1.1.0\n",
      "tf-slim                      1.1.0\n",
      "tifffile                     2023.12.9\n",
      "tinycss2                     1.2.1\n",
      "torch                        2.1.0+cu118\n",
      "torchaudio                   2.1.0+cu118\n",
      "torchvision                  0.16.0+cu118\n",
      "tornado                      6.4\n",
      "traitlets                    5.14.1\n",
      "types-python-dateutil        2.8.19.20240106\n",
      "typing_extensions            4.5.0\n",
      "uri-template                 1.3.0\n",
      "urllib3                      2.1.0\n",
      "wcwidth                      0.2.13\n",
      "webcolors                    1.13\n",
      "webencodings                 0.5.1\n",
      "websocket-client             1.7.0\n",
      "Werkzeug                     3.0.1\n",
      "wget                         3.2\n",
      "wheel                        0.42.0\n",
      "widgetsnbextension           4.0.9\n",
      "wrapt                        1.14.1\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING: Ignoring invalid distribution ~atplotlib (C:\\Users\\Hasitha\\Desktop\\ANPR\\anprsys\\Lib\\site-packages)\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\absl_py-2.0.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\apache_beam-2.53.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\avro_python3-1.10.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\contextlib2-21.6.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\cython-3.0.8-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\dm_tree-0.1.8-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\h5py-3.10.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lvis-0.5.3-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lxml-5.1.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\namex-0.0.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\object_detection-0.1-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pandas-2.2.0rc0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pillow-10.2.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\portalocker-2.8.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pycocotools-2.0.7-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pyparsing-2.4.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\regex-2023.12.25-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\rich-13.7.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\sacrebleu-2.2.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\scipy-1.12.0rc2-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tabulate-0.9.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io_gcs_filesystem-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_models_official-2.15.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_slim-1.1.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n"
     ]
    }
   ],
   "source": [
    "pip list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Running tests under Python 3.11.4: C:\\Users\\Hasitha\\Desktop\\ANPR\\anprsys\\Scripts\\python.exe\n",
      "[ RUN      ] ModelBuilderTF2Test.test_create_center_net_deepmac\n",
      "2024-01-20 09:08:14.637825: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n",
      "To enable the following instructions: SSE SSE2 SSE3 SSE4.1 SSE4.2 AVX AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
      "WARNING:tensorflow:`tf.keras.layers.experimental.SyncBatchNormalization` endpoint is deprecated and will be removed in a future release. Please use `tf.keras.layers.BatchNormalization` with parameter `synchronized` set to True.\n",
      "W0120 09:08:14.694226   320 batch_normalization.py:1531] `tf.keras.layers.experimental.SyncBatchNormalization` endpoint is deprecated and will be removed in a future release. Please use `tf.keras.layers.BatchNormalization` with parameter `synchronized` set to True.\n",
      "C:\\Users\\Hasitha\\Desktop\\ANPR\\anprsys\\Lib\\site-packages\\object_detection-0.1-py3.11.egg\\object_detection\\builders\\model_builder.py:1112: DeprecationWarning: The 'warn' function is deprecated, use 'warning' instead\n",
      "  logging.warn(('Building experimental DeepMAC meta-arch.'\n",
      "W0120 09:08:14.901225   320 model_builder.py:1112] Building experimental DeepMAC meta-arch. Some features may be omitted.\n",
      "INFO:tensorflow:time(__main__.ModelBuilderTF2Test.test_create_center_net_deepmac): 0.52s\n",
      "I0120 09:08:15.144549   320 test_util.py:2462] time(__main__.ModelBuilderTF2Test.test_create_center_net_deepmac): 0.52s\n",
      "[       OK ] ModelBuilderTF2Test.test_create_center_net_deepmac\n",
      "[ RUN      ] ModelBuilderTF2Test.test_create_center_net_model0 (customize_head_params=True)\n",
      "INFO:tensorflow:time(__main__.ModelBuilderTF2Test.test_create_center_net_model0 (customize_head_params=True)): 0.88s\n",
      "I0120 09:08:16.024919   320 test_util.py:2462] time(__main__.ModelBuilderTF2Test.test_create_center_net_model0 (customize_head_params=True)): 0.88s\n",
      "[       OK ] ModelBuilderTF2Test.test_create_center_net_model0 (customize_head_params=True)\n",
      "[ RUN      ] ModelBuilderTF2Test.test_create_center_net_model1 (customize_head_params=False)\n",
      "INFO:tensorflow:time(__main__.ModelBuilderTF2Test.test_create_center_net_model1 (customize_head_params=False)): 0.24s\n",
      "I0120 09:08:16.264284   320 test_util.py:2462] time(__main__.ModelBuilderTF2Test.test_create_center_net_model1 (customize_head_params=False)): 0.24s\n",
      "[       OK ] ModelBuilderTF2Test.test_create_center_net_model1 (customize_head_params=False)\n",
      "[ RUN      ] ModelBuilderTF2Test.test_create_center_net_model_from_keypoints\n",
      "INFO:tensorflow:time(__main__.ModelBuilderTF2Test.test_create_center_net_model_from_keypoints): 0.22s\n",
      "I0120 09:08:16.484607   320 test_util.py:2462] time(__main__.ModelBuilderTF2Test.test_create_center_net_model_from_keypoints): 0.22s\n",
      "[       OK ] ModelBuilderTF2Test.test_create_center_net_model_from_keypoints\n",
      "[ RUN      ] ModelBuilderTF2Test.test_create_center_net_model_mobilenet\n",
      "INFO:tensorflow:time(__main__.ModelBuilderTF2Test.test_create_center_net_model_mobilenet): 2.07s\n",
      "I0120 09:08:18.552793   320 test_util.py:2462] time(__main__.ModelBuilderTF2Test.test_create_center_net_model_mobilenet): 2.07s\n",
      "[       OK ] ModelBuilderTF2Test.test_create_center_net_model_mobilenet\n",
      "[ RUN      ] ModelBuilderTF2Test.test_create_experimental_model\n",
      "INFO:tensorflow:time(__main__.ModelBuilderTF2Test.test_create_experimental_model): 0.0s\n",
      "I0120 09:08:18.557114   320 test_util.py:2462] time(__main__.ModelBuilderTF2Test.test_create_experimental_model): 0.0s\n",
      "[       OK ] ModelBuilderTF2Test.test_create_experimental_model\n",
      "[ RUN      ] ModelBuilderTF2Test.test_create_faster_rcnn_from_config_with_crop_feature0 (True)\n",
      "INFO:tensorflow:time(__main__.ModelBuilderTF2Test.test_create_faster_rcnn_from_config_with_crop_feature0 (True)): 0.02s\n",
      "I0120 09:08:18.578656   320 test_util.py:2462] time(__main__.ModelBuilderTF2Test.test_create_faster_rcnn_from_config_with_crop_feature0 (True)): 0.02s\n",
      "[       OK ] ModelBuilderTF2Test.test_create_faster_rcnn_from_config_with_crop_feature0 (True)\n",
      "[ RUN      ] ModelBuilderTF2Test.test_create_faster_rcnn_from_config_with_crop_feature1 (False)\n",
      "INFO:tensorflow:time(__main__.ModelBuilderTF2Test.test_create_faster_rcnn_from_config_with_crop_feature1 (False)): 0.01s\n",
      "I0120 09:08:18.591171   320 test_util.py:2462] time(__main__.ModelBuilderTF2Test.test_create_faster_rcnn_from_config_with_crop_feature1 (False)): 0.01s\n",
      "[       OK ] ModelBuilderTF2Test.test_create_faster_rcnn_from_config_with_crop_feature1 (False)\n",
      "[ RUN      ] ModelBuilderTF2Test.test_create_faster_rcnn_model_from_config_with_example_miner\n",
      "INFO:tensorflow:time(__main__.ModelBuilderTF2Test.test_create_faster_rcnn_model_from_config_with_example_miner): 0.01s\n",
      "I0120 09:08:18.606190   320 test_util.py:2462] time(__main__.ModelBuilderTF2Test.test_create_faster_rcnn_model_from_config_with_example_miner): 0.01s\n",
      "[       OK ] ModelBuilderTF2Test.test_create_faster_rcnn_model_from_config_with_example_miner\n",
      "[ RUN      ] ModelBuilderTF2Test.test_create_faster_rcnn_models_from_config_faster_rcnn_with_matmul\n",
      "INFO:tensorflow:time(__main__.ModelBuilderTF2Test.test_create_faster_rcnn_models_from_config_faster_rcnn_with_matmul): 0.08s\n",
      "I0120 09:08:18.686135   320 test_util.py:2462] time(__main__.ModelBuilderTF2Test.test_create_faster_rcnn_models_from_config_faster_rcnn_with_matmul): 0.08s\n",
      "[       OK ] ModelBuilderTF2Test.test_create_faster_rcnn_models_from_config_faster_rcnn_with_matmul\n",
      "[ RUN      ] ModelBuilderTF2Test.test_create_faster_rcnn_models_from_config_faster_rcnn_without_matmul\n",
      "INFO:tensorflow:time(__main__.ModelBuilderTF2Test.test_create_faster_rcnn_models_from_config_faster_rcnn_without_matmul): 0.08s\n",
      "I0120 09:08:18.764646   320 test_util.py:2462] time(__main__.ModelBuilderTF2Test.test_create_faster_rcnn_models_from_config_faster_rcnn_without_matmul): 0.08s\n",
      "[       OK ] ModelBuilderTF2Test.test_create_faster_rcnn_models_from_config_faster_rcnn_without_matmul\n",
      "[ RUN      ] ModelBuilderTF2Test.test_create_faster_rcnn_models_from_config_mask_rcnn_with_matmul\n",
      "INFO:tensorflow:time(__main__.ModelBuilderTF2Test.test_create_faster_rcnn_models_from_config_mask_rcnn_with_matmul): 0.08s\n",
      "I0120 09:08:18.847698   320 test_util.py:2462] time(__main__.ModelBuilderTF2Test.test_create_faster_rcnn_models_from_config_mask_rcnn_with_matmul): 0.08s\n",
      "[       OK ] ModelBuilderTF2Test.test_create_faster_rcnn_models_from_config_mask_rcnn_with_matmul\n",
      "[ RUN      ] ModelBuilderTF2Test.test_create_faster_rcnn_models_from_config_mask_rcnn_without_matmul\n",
      "INFO:tensorflow:time(__main__.ModelBuilderTF2Test.test_create_faster_rcnn_models_from_config_mask_rcnn_without_matmul): 0.08s\n",
      "I0120 09:08:18.929383   320 test_util.py:2462] time(__main__.ModelBuilderTF2Test.test_create_faster_rcnn_models_from_config_mask_rcnn_without_matmul): 0.08s\n",
      "[       OK ] ModelBuilderTF2Test.test_create_faster_rcnn_models_from_config_mask_rcnn_without_matmul\n",
      "[ RUN      ] ModelBuilderTF2Test.test_create_rfcn_model_from_config\n",
      "INFO:tensorflow:time(__main__.ModelBuilderTF2Test.test_create_rfcn_model_from_config): 0.08s\n",
      "I0120 09:08:19.010810   320 test_util.py:2462] time(__main__.ModelBuilderTF2Test.test_create_rfcn_model_from_config): 0.08s\n",
      "[       OK ] ModelBuilderTF2Test.test_create_rfcn_model_from_config\n",
      "[ RUN      ] ModelBuilderTF2Test.test_create_ssd_fpn_model_from_config\n",
      "INFO:tensorflow:time(__main__.ModelBuilderTF2Test.test_create_ssd_fpn_model_from_config): 0.02s\n",
      "I0120 09:08:19.034358   320 test_util.py:2462] time(__main__.ModelBuilderTF2Test.test_create_ssd_fpn_model_from_config): 0.02s\n",
      "[       OK ] ModelBuilderTF2Test.test_create_ssd_fpn_model_from_config\n",
      "[ RUN      ] ModelBuilderTF2Test.test_create_ssd_models_from_config\n",
      "I0120 09:08:19.178635   320 ssd_efficientnet_bifpn_feature_extractor.py:161] EfficientDet EfficientNet backbone version: efficientnet-b0\n",
      "I0120 09:08:19.178635   320 ssd_efficientnet_bifpn_feature_extractor.py:163] EfficientDet BiFPN num filters: 64\n",
      "I0120 09:08:19.179637   320 ssd_efficientnet_bifpn_feature_extractor.py:164] EfficientDet BiFPN num iterations: 3\n",
      "I0120 09:08:19.181140   320 efficientnet_model.py:143] round_filter input=32 output=32\n",
      "I0120 09:08:19.203812   320 efficientnet_model.py:143] round_filter input=32 output=32\n",
      "I0120 09:08:19.203812   320 efficientnet_model.py:143] round_filter input=16 output=16\n",
      "I0120 09:08:19.275074   320 efficientnet_model.py:143] round_filter input=16 output=16\n",
      "I0120 09:08:19.275074   320 efficientnet_model.py:143] round_filter input=24 output=24\n",
      "I0120 09:08:19.448397   320 efficientnet_model.py:143] round_filter input=24 output=24\n",
      "I0120 09:08:19.448397   320 efficientnet_model.py:143] round_filter input=40 output=40\n",
      "I0120 09:08:19.618589   320 efficientnet_model.py:143] round_filter input=40 output=40\n",
      "I0120 09:08:19.618589   320 efficientnet_model.py:143] round_filter input=80 output=80\n",
      "I0120 09:08:19.871462   320 efficientnet_model.py:143] round_filter input=80 output=80\n",
      "I0120 09:08:19.871462   320 efficientnet_model.py:143] round_filter input=112 output=112\n",
      "I0120 09:08:20.135370   320 efficientnet_model.py:143] round_filter input=112 output=112\n",
      "I0120 09:08:20.135370   320 efficientnet_model.py:143] round_filter input=192 output=192\n",
      "I0120 09:08:20.493688   320 efficientnet_model.py:143] round_filter input=192 output=192\n",
      "I0120 09:08:20.493688   320 efficientnet_model.py:143] round_filter input=320 output=320\n",
      "I0120 09:08:20.590727   320 efficientnet_model.py:143] round_filter input=1280 output=1280\n",
      "I0120 09:08:20.644875   320 efficientnet_model.py:453] Building model efficientnet with params ModelConfig(width_coefficient=1.0, depth_coefficient=1.0, resolution=224, dropout_rate=0.2, blocks=(BlockConfig(input_filters=32, output_filters=16, kernel_size=3, num_repeat=1, expand_ratio=1, strides=(1, 1), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=16, output_filters=24, kernel_size=3, num_repeat=2, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=24, output_filters=40, kernel_size=5, num_repeat=2, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=40, output_filters=80, kernel_size=3, num_repeat=3, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=80, output_filters=112, kernel_size=5, num_repeat=3, expand_ratio=6, strides=(1, 1), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=112, output_filters=192, kernel_size=5, num_repeat=4, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=192, output_filters=320, kernel_size=3, num_repeat=1, expand_ratio=6, strides=(1, 1), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise')), stem_base_filters=32, top_base_filters=1280, activation='simple_swish', batch_norm='default', bn_momentum=0.99, bn_epsilon=0.001, weight_decay=5e-06, drop_connect_rate=0.2, depth_divisor=8, min_depth=None, use_se=True, input_channels=3, num_classes=1000, model_name='efficientnet', rescale_input=False, data_format='channels_last', dtype='float32')\n",
      "I0120 09:08:20.688022   320 ssd_efficientnet_bifpn_feature_extractor.py:161] EfficientDet EfficientNet backbone version: efficientnet-b1\n",
      "I0120 09:08:20.688022   320 ssd_efficientnet_bifpn_feature_extractor.py:163] EfficientDet BiFPN num filters: 88\n",
      "I0120 09:08:20.688022   320 ssd_efficientnet_bifpn_feature_extractor.py:164] EfficientDet BiFPN num iterations: 4\n",
      "I0120 09:08:20.689022   320 efficientnet_model.py:143] round_filter input=32 output=32\n",
      "I0120 09:08:20.705272   320 efficientnet_model.py:143] round_filter input=32 output=32\n",
      "I0120 09:08:20.705272   320 efficientnet_model.py:143] round_filter input=16 output=16\n",
      "I0120 09:08:20.833170   320 efficientnet_model.py:143] round_filter input=16 output=16\n",
      "I0120 09:08:20.833170   320 efficientnet_model.py:143] round_filter input=24 output=24\n",
      "I0120 09:08:21.073908   320 efficientnet_model.py:143] round_filter input=24 output=24\n",
      "I0120 09:08:21.073908   320 efficientnet_model.py:143] round_filter input=40 output=40\n",
      "I0120 09:08:21.316988   320 efficientnet_model.py:143] round_filter input=40 output=40\n",
      "I0120 09:08:21.318255   320 efficientnet_model.py:143] round_filter input=80 output=80\n",
      "I0120 09:08:21.645315   320 efficientnet_model.py:143] round_filter input=80 output=80\n",
      "I0120 09:08:21.645315   320 efficientnet_model.py:143] round_filter input=112 output=112\n",
      "I0120 09:08:22.163433   320 efficientnet_model.py:143] round_filter input=112 output=112\n",
      "I0120 09:08:22.164433   320 efficientnet_model.py:143] round_filter input=192 output=192\n",
      "I0120 09:08:22.582138   320 efficientnet_model.py:143] round_filter input=192 output=192\n",
      "I0120 09:08:22.582138   320 efficientnet_model.py:143] round_filter input=320 output=320\n",
      "I0120 09:08:22.761529   320 efficientnet_model.py:143] round_filter input=1280 output=1280\n",
      "I0120 09:08:22.798470   320 efficientnet_model.py:453] Building model efficientnet with params ModelConfig(width_coefficient=1.0, depth_coefficient=1.1, resolution=240, dropout_rate=0.2, blocks=(BlockConfig(input_filters=32, output_filters=16, kernel_size=3, num_repeat=1, expand_ratio=1, strides=(1, 1), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=16, output_filters=24, kernel_size=3, num_repeat=2, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=24, output_filters=40, kernel_size=5, num_repeat=2, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=40, output_filters=80, kernel_size=3, num_repeat=3, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=80, output_filters=112, kernel_size=5, num_repeat=3, expand_ratio=6, strides=(1, 1), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=112, output_filters=192, kernel_size=5, num_repeat=4, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=192, output_filters=320, kernel_size=3, num_repeat=1, expand_ratio=6, strides=(1, 1), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise')), stem_base_filters=32, top_base_filters=1280, activation='simple_swish', batch_norm='default', bn_momentum=0.99, bn_epsilon=0.001, weight_decay=5e-06, drop_connect_rate=0.2, depth_divisor=8, min_depth=None, use_se=True, input_channels=3, num_classes=1000, model_name='efficientnet', rescale_input=False, data_format='channels_last', dtype='float32')\n",
      "I0120 09:08:22.849401   320 ssd_efficientnet_bifpn_feature_extractor.py:161] EfficientDet EfficientNet backbone version: efficientnet-b2\n",
      "I0120 09:08:22.849401   320 ssd_efficientnet_bifpn_feature_extractor.py:163] EfficientDet BiFPN num filters: 112\n",
      "I0120 09:08:22.849401   320 ssd_efficientnet_bifpn_feature_extractor.py:164] EfficientDet BiFPN num iterations: 5\n",
      "I0120 09:08:22.850906   320 efficientnet_model.py:143] round_filter input=32 output=32\n",
      "I0120 09:08:22.866602   320 efficientnet_model.py:143] round_filter input=32 output=32\n",
      "I0120 09:08:22.866602   320 efficientnet_model.py:143] round_filter input=16 output=16\n",
      "I0120 09:08:22.988648   320 efficientnet_model.py:143] round_filter input=16 output=16\n",
      "I0120 09:08:22.988648   320 efficientnet_model.py:143] round_filter input=24 output=24\n",
      "I0120 09:08:23.224332   320 efficientnet_model.py:143] round_filter input=24 output=24\n",
      "I0120 09:08:23.224332   320 efficientnet_model.py:143] round_filter input=40 output=48\n",
      "I0120 09:08:23.473356   320 efficientnet_model.py:143] round_filter input=40 output=48\n",
      "I0120 09:08:23.473356   320 efficientnet_model.py:143] round_filter input=80 output=88\n",
      "I0120 09:08:23.807417   320 efficientnet_model.py:143] round_filter input=80 output=88\n",
      "I0120 09:08:23.807417   320 efficientnet_model.py:143] round_filter input=112 output=120\n",
      "I0120 09:08:24.142094   320 efficientnet_model.py:143] round_filter input=112 output=120\n",
      "I0120 09:08:24.142094   320 efficientnet_model.py:143] round_filter input=192 output=208\n",
      "I0120 09:08:24.569966   320 efficientnet_model.py:143] round_filter input=192 output=208\n",
      "I0120 09:08:24.569966   320 efficientnet_model.py:143] round_filter input=320 output=352\n",
      "I0120 09:08:24.764850   320 efficientnet_model.py:143] round_filter input=1280 output=1408\n",
      "I0120 09:08:24.807848   320 efficientnet_model.py:453] Building model efficientnet with params ModelConfig(width_coefficient=1.1, depth_coefficient=1.2, resolution=260, dropout_rate=0.3, blocks=(BlockConfig(input_filters=32, output_filters=16, kernel_size=3, num_repeat=1, expand_ratio=1, strides=(1, 1), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=16, output_filters=24, kernel_size=3, num_repeat=2, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=24, output_filters=40, kernel_size=5, num_repeat=2, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=40, output_filters=80, kernel_size=3, num_repeat=3, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=80, output_filters=112, kernel_size=5, num_repeat=3, expand_ratio=6, strides=(1, 1), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=112, output_filters=192, kernel_size=5, num_repeat=4, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=192, output_filters=320, kernel_size=3, num_repeat=1, expand_ratio=6, strides=(1, 1), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise')), stem_base_filters=32, top_base_filters=1280, activation='simple_swish', batch_norm='default', bn_momentum=0.99, bn_epsilon=0.001, weight_decay=5e-06, drop_connect_rate=0.2, depth_divisor=8, min_depth=None, use_se=True, input_channels=3, num_classes=1000, model_name='efficientnet', rescale_input=False, data_format='channels_last', dtype='float32')\n",
      "I0120 09:08:24.858386   320 ssd_efficientnet_bifpn_feature_extractor.py:161] EfficientDet EfficientNet backbone version: efficientnet-b3\n",
      "I0120 09:08:24.858386   320 ssd_efficientnet_bifpn_feature_extractor.py:163] EfficientDet BiFPN num filters: 160\n",
      "I0120 09:08:24.858386   320 ssd_efficientnet_bifpn_feature_extractor.py:164] EfficientDet BiFPN num iterations: 6\n",
      "I0120 09:08:24.860391   320 efficientnet_model.py:143] round_filter input=32 output=40\n",
      "I0120 09:08:24.877900   320 efficientnet_model.py:143] round_filter input=32 output=40\n",
      "I0120 09:08:24.877900   320 efficientnet_model.py:143] round_filter input=16 output=24\n",
      "I0120 09:08:25.013214   320 efficientnet_model.py:143] round_filter input=16 output=24\n",
      "I0120 09:08:25.013214   320 efficientnet_model.py:143] round_filter input=24 output=32\n",
      "I0120 09:08:25.255258   320 efficientnet_model.py:143] round_filter input=24 output=32\n",
      "I0120 09:08:25.255258   320 efficientnet_model.py:143] round_filter input=40 output=48\n",
      "I0120 09:08:25.497074   320 efficientnet_model.py:143] round_filter input=40 output=48\n",
      "I0120 09:08:25.497074   320 efficientnet_model.py:143] round_filter input=80 output=96\n",
      "I0120 09:08:25.908382   320 efficientnet_model.py:143] round_filter input=80 output=96\n",
      "I0120 09:08:25.909387   320 efficientnet_model.py:143] round_filter input=112 output=136\n",
      "I0120 09:08:26.328567   320 efficientnet_model.py:143] round_filter input=112 output=136\n",
      "I0120 09:08:26.329566   320 efficientnet_model.py:143] round_filter input=192 output=232\n",
      "I0120 09:08:26.857576   320 efficientnet_model.py:143] round_filter input=192 output=232\n",
      "I0120 09:08:26.857576   320 efficientnet_model.py:143] round_filter input=320 output=384\n",
      "I0120 09:08:27.050778   320 efficientnet_model.py:143] round_filter input=1280 output=1536\n",
      "I0120 09:08:27.095328   320 efficientnet_model.py:453] Building model efficientnet with params ModelConfig(width_coefficient=1.2, depth_coefficient=1.4, resolution=300, dropout_rate=0.3, blocks=(BlockConfig(input_filters=32, output_filters=16, kernel_size=3, num_repeat=1, expand_ratio=1, strides=(1, 1), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=16, output_filters=24, kernel_size=3, num_repeat=2, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=24, output_filters=40, kernel_size=5, num_repeat=2, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=40, output_filters=80, kernel_size=3, num_repeat=3, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=80, output_filters=112, kernel_size=5, num_repeat=3, expand_ratio=6, strides=(1, 1), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=112, output_filters=192, kernel_size=5, num_repeat=4, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=192, output_filters=320, kernel_size=3, num_repeat=1, expand_ratio=6, strides=(1, 1), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise')), stem_base_filters=32, top_base_filters=1280, activation='simple_swish', batch_norm='default', bn_momentum=0.99, bn_epsilon=0.001, weight_decay=5e-06, drop_connect_rate=0.2, depth_divisor=8, min_depth=None, use_se=True, input_channels=3, num_classes=1000, model_name='efficientnet', rescale_input=False, data_format='channels_last', dtype='float32')\n",
      "I0120 09:08:27.150967   320 ssd_efficientnet_bifpn_feature_extractor.py:161] EfficientDet EfficientNet backbone version: efficientnet-b4\n",
      "I0120 09:08:27.150967   320 ssd_efficientnet_bifpn_feature_extractor.py:163] EfficientDet BiFPN num filters: 224\n",
      "I0120 09:08:27.150967   320 ssd_efficientnet_bifpn_feature_extractor.py:164] EfficientDet BiFPN num iterations: 7\n",
      "I0120 09:08:27.151967   320 efficientnet_model.py:143] round_filter input=32 output=48\n",
      "I0120 09:08:27.169978   320 efficientnet_model.py:143] round_filter input=32 output=48\n",
      "I0120 09:08:27.169978   320 efficientnet_model.py:143] round_filter input=16 output=24\n",
      "I0120 09:08:27.298085   320 efficientnet_model.py:143] round_filter input=16 output=24\n",
      "I0120 09:08:27.298085   320 efficientnet_model.py:143] round_filter input=24 output=32\n",
      "I0120 09:08:27.611940   320 efficientnet_model.py:143] round_filter input=24 output=32\n",
      "I0120 09:08:27.611940   320 efficientnet_model.py:143] round_filter input=40 output=56\n",
      "I0120 09:08:28.188481   320 efficientnet_model.py:143] round_filter input=40 output=56\n",
      "I0120 09:08:28.188481   320 efficientnet_model.py:143] round_filter input=80 output=112\n",
      "I0120 09:08:28.680308   320 efficientnet_model.py:143] round_filter input=80 output=112\n",
      "I0120 09:08:28.680308   320 efficientnet_model.py:143] round_filter input=112 output=160\n",
      "I0120 09:08:29.184194   320 efficientnet_model.py:143] round_filter input=112 output=160\n",
      "I0120 09:08:29.184194   320 efficientnet_model.py:143] round_filter input=192 output=272\n",
      "I0120 09:08:29.900202   320 efficientnet_model.py:143] round_filter input=192 output=272\n",
      "I0120 09:08:29.900202   320 efficientnet_model.py:143] round_filter input=320 output=448\n",
      "I0120 09:08:30.104239   320 efficientnet_model.py:143] round_filter input=1280 output=1792\n",
      "I0120 09:08:30.151770   320 efficientnet_model.py:453] Building model efficientnet with params ModelConfig(width_coefficient=1.4, depth_coefficient=1.8, resolution=380, dropout_rate=0.4, blocks=(BlockConfig(input_filters=32, output_filters=16, kernel_size=3, num_repeat=1, expand_ratio=1, strides=(1, 1), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=16, output_filters=24, kernel_size=3, num_repeat=2, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=24, output_filters=40, kernel_size=5, num_repeat=2, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=40, output_filters=80, kernel_size=3, num_repeat=3, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=80, output_filters=112, kernel_size=5, num_repeat=3, expand_ratio=6, strides=(1, 1), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=112, output_filters=192, kernel_size=5, num_repeat=4, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=192, output_filters=320, kernel_size=3, num_repeat=1, expand_ratio=6, strides=(1, 1), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise')), stem_base_filters=32, top_base_filters=1280, activation='simple_swish', batch_norm='default', bn_momentum=0.99, bn_epsilon=0.001, weight_decay=5e-06, drop_connect_rate=0.2, depth_divisor=8, min_depth=None, use_se=True, input_channels=3, num_classes=1000, model_name='efficientnet', rescale_input=False, data_format='channels_last', dtype='float32')\n",
      "I0120 09:08:30.215641   320 ssd_efficientnet_bifpn_feature_extractor.py:161] EfficientDet EfficientNet backbone version: efficientnet-b5\n",
      "I0120 09:08:30.215641   320 ssd_efficientnet_bifpn_feature_extractor.py:163] EfficientDet BiFPN num filters: 288\n",
      "I0120 09:08:30.215641   320 ssd_efficientnet_bifpn_feature_extractor.py:164] EfficientDet BiFPN num iterations: 7\n",
      "I0120 09:08:30.217151   320 efficientnet_model.py:143] round_filter input=32 output=48\n",
      "I0120 09:08:30.232835   320 efficientnet_model.py:143] round_filter input=32 output=48\n",
      "I0120 09:08:30.232835   320 efficientnet_model.py:143] round_filter input=16 output=24\n",
      "I0120 09:08:30.417418   320 efficientnet_model.py:143] round_filter input=16 output=24\n",
      "I0120 09:08:30.417418   320 efficientnet_model.py:143] round_filter input=24 output=40\n",
      "I0120 09:08:30.817727   320 efficientnet_model.py:143] round_filter input=24 output=40\n",
      "I0120 09:08:30.817727   320 efficientnet_model.py:143] round_filter input=40 output=64\n",
      "I0120 09:08:31.232162   320 efficientnet_model.py:143] round_filter input=40 output=64\n",
      "I0120 09:08:31.232162   320 efficientnet_model.py:143] round_filter input=80 output=128\n",
      "I0120 09:08:31.818659   320 efficientnet_model.py:143] round_filter input=80 output=128\n",
      "I0120 09:08:31.818659   320 efficientnet_model.py:143] round_filter input=112 output=176\n",
      "I0120 09:08:32.407339   320 efficientnet_model.py:143] round_filter input=112 output=176\n",
      "I0120 09:08:32.407339   320 efficientnet_model.py:143] round_filter input=192 output=304\n",
      "I0120 09:08:33.224853   320 efficientnet_model.py:143] round_filter input=192 output=304\n",
      "I0120 09:08:33.224853   320 efficientnet_model.py:143] round_filter input=320 output=512\n",
      "I0120 09:08:33.538305   320 efficientnet_model.py:143] round_filter input=1280 output=2048\n",
      "I0120 09:08:33.586619   320 efficientnet_model.py:453] Building model efficientnet with params ModelConfig(width_coefficient=1.6, depth_coefficient=2.2, resolution=456, dropout_rate=0.4, blocks=(BlockConfig(input_filters=32, output_filters=16, kernel_size=3, num_repeat=1, expand_ratio=1, strides=(1, 1), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=16, output_filters=24, kernel_size=3, num_repeat=2, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=24, output_filters=40, kernel_size=5, num_repeat=2, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=40, output_filters=80, kernel_size=3, num_repeat=3, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=80, output_filters=112, kernel_size=5, num_repeat=3, expand_ratio=6, strides=(1, 1), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=112, output_filters=192, kernel_size=5, num_repeat=4, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=192, output_filters=320, kernel_size=3, num_repeat=1, expand_ratio=6, strides=(1, 1), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise')), stem_base_filters=32, top_base_filters=1280, activation='simple_swish', batch_norm='default', bn_momentum=0.99, bn_epsilon=0.001, weight_decay=5e-06, drop_connect_rate=0.2, depth_divisor=8, min_depth=None, use_se=True, input_channels=3, num_classes=1000, model_name='efficientnet', rescale_input=False, data_format='channels_last', dtype='float32')\n",
      "I0120 09:08:33.662309   320 ssd_efficientnet_bifpn_feature_extractor.py:161] EfficientDet EfficientNet backbone version: efficientnet-b6\n",
      "I0120 09:08:33.662309   320 ssd_efficientnet_bifpn_feature_extractor.py:163] EfficientDet BiFPN num filters: 384\n",
      "I0120 09:08:33.662309   320 ssd_efficientnet_bifpn_feature_extractor.py:164] EfficientDet BiFPN num iterations: 8\n",
      "I0120 09:08:33.663314   320 efficientnet_model.py:143] round_filter input=32 output=56\n",
      "I0120 09:08:33.680121   320 efficientnet_model.py:143] round_filter input=32 output=56\n",
      "I0120 09:08:33.681121   320 efficientnet_model.py:143] round_filter input=16 output=32\n",
      "I0120 09:08:33.874156   320 efficientnet_model.py:143] round_filter input=16 output=32\n",
      "I0120 09:08:33.874156   320 efficientnet_model.py:143] round_filter input=24 output=40\n",
      "I0120 09:08:34.350405   320 efficientnet_model.py:143] round_filter input=24 output=40\n",
      "I0120 09:08:34.350405   320 efficientnet_model.py:143] round_filter input=40 output=72\n",
      "I0120 09:08:35.115727   320 efficientnet_model.py:143] round_filter input=40 output=72\n",
      "I0120 09:08:35.115727   320 efficientnet_model.py:143] round_filter input=80 output=144\n",
      "I0120 09:08:35.833274   320 efficientnet_model.py:143] round_filter input=80 output=144\n",
      "I0120 09:08:35.833274   320 efficientnet_model.py:143] round_filter input=112 output=200\n",
      "I0120 09:08:36.519582   320 efficientnet_model.py:143] round_filter input=112 output=200\n",
      "I0120 09:08:36.519582   320 efficientnet_model.py:143] round_filter input=192 output=344\n",
      "I0120 09:08:37.549622   320 efficientnet_model.py:143] round_filter input=192 output=344\n",
      "I0120 09:08:37.549622   320 efficientnet_model.py:143] round_filter input=320 output=576\n",
      "I0120 09:08:37.871815   320 efficientnet_model.py:143] round_filter input=1280 output=2304\n",
      "I0120 09:08:37.921574   320 efficientnet_model.py:453] Building model efficientnet with params ModelConfig(width_coefficient=1.8, depth_coefficient=2.6, resolution=528, dropout_rate=0.5, blocks=(BlockConfig(input_filters=32, output_filters=16, kernel_size=3, num_repeat=1, expand_ratio=1, strides=(1, 1), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=16, output_filters=24, kernel_size=3, num_repeat=2, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=24, output_filters=40, kernel_size=5, num_repeat=2, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=40, output_filters=80, kernel_size=3, num_repeat=3, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=80, output_filters=112, kernel_size=5, num_repeat=3, expand_ratio=6, strides=(1, 1), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=112, output_filters=192, kernel_size=5, num_repeat=4, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=192, output_filters=320, kernel_size=3, num_repeat=1, expand_ratio=6, strides=(1, 1), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise')), stem_base_filters=32, top_base_filters=1280, activation='simple_swish', batch_norm='default', bn_momentum=0.99, bn_epsilon=0.001, weight_decay=5e-06, drop_connect_rate=0.2, depth_divisor=8, min_depth=None, use_se=True, input_channels=3, num_classes=1000, model_name='efficientnet', rescale_input=False, data_format='channels_last', dtype='float32')\n",
      "I0120 09:08:38.008039   320 ssd_efficientnet_bifpn_feature_extractor.py:161] EfficientDet EfficientNet backbone version: efficientnet-b7\n",
      "I0120 09:08:38.008039   320 ssd_efficientnet_bifpn_feature_extractor.py:163] EfficientDet BiFPN num filters: 384\n",
      "I0120 09:08:38.008039   320 ssd_efficientnet_bifpn_feature_extractor.py:164] EfficientDet BiFPN num iterations: 8\n",
      "I0120 09:08:38.010039   320 efficientnet_model.py:143] round_filter input=32 output=64\n",
      "I0120 09:08:38.028098   320 efficientnet_model.py:143] round_filter input=32 output=64\n",
      "I0120 09:08:38.028098   320 efficientnet_model.py:143] round_filter input=16 output=32\n",
      "I0120 09:08:38.285194   320 efficientnet_model.py:143] round_filter input=16 output=32\n",
      "I0120 09:08:38.285194   320 efficientnet_model.py:143] round_filter input=24 output=48\n",
      "I0120 09:08:38.847315   320 efficientnet_model.py:143] round_filter input=24 output=48\n",
      "I0120 09:08:38.847315   320 efficientnet_model.py:143] round_filter input=40 output=80\n",
      "I0120 09:08:39.413868   320 efficientnet_model.py:143] round_filter input=40 output=80\n",
      "I0120 09:08:39.413868   320 efficientnet_model.py:143] round_filter input=80 output=160\n",
      "I0120 09:08:40.249945   320 efficientnet_model.py:143] round_filter input=80 output=160\n",
      "I0120 09:08:40.249945   320 efficientnet_model.py:143] round_filter input=112 output=224\n",
      "I0120 09:08:41.120833   320 efficientnet_model.py:143] round_filter input=112 output=224\n",
      "I0120 09:08:41.120833   320 efficientnet_model.py:143] round_filter input=192 output=384\n",
      "I0120 09:08:42.714425   320 efficientnet_model.py:143] round_filter input=192 output=384\n",
      "I0120 09:08:42.715426   320 efficientnet_model.py:143] round_filter input=320 output=640\n",
      "I0120 09:08:43.173358   320 efficientnet_model.py:143] round_filter input=1280 output=2560\n",
      "I0120 09:08:43.227068   320 efficientnet_model.py:453] Building model efficientnet with params ModelConfig(width_coefficient=2.0, depth_coefficient=3.1, resolution=600, dropout_rate=0.5, blocks=(BlockConfig(input_filters=32, output_filters=16, kernel_size=3, num_repeat=1, expand_ratio=1, strides=(1, 1), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=16, output_filters=24, kernel_size=3, num_repeat=2, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=24, output_filters=40, kernel_size=5, num_repeat=2, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=40, output_filters=80, kernel_size=3, num_repeat=3, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=80, output_filters=112, kernel_size=5, num_repeat=3, expand_ratio=6, strides=(1, 1), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=112, output_filters=192, kernel_size=5, num_repeat=4, expand_ratio=6, strides=(2, 2), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise'), BlockConfig(input_filters=192, output_filters=320, kernel_size=3, num_repeat=1, expand_ratio=6, strides=(1, 1), se_ratio=0.25, id_skip=True, fused_conv=False, conv_type='depthwise')), stem_base_filters=32, top_base_filters=1280, activation='simple_swish', batch_norm='default', bn_momentum=0.99, bn_epsilon=0.001, weight_decay=5e-06, drop_connect_rate=0.2, depth_divisor=8, min_depth=None, use_se=True, input_channels=3, num_classes=1000, model_name='efficientnet', rescale_input=False, data_format='channels_last', dtype='float32')\n",
      "INFO:tensorflow:time(__main__.ModelBuilderTF2Test.test_create_ssd_models_from_config): 24.29s\n",
      "I0120 09:08:43.325642   320 test_util.py:2462] time(__main__.ModelBuilderTF2Test.test_create_ssd_models_from_config): 24.29s\n",
      "[       OK ] ModelBuilderTF2Test.test_create_ssd_models_from_config\n",
      "[ RUN      ] ModelBuilderTF2Test.test_invalid_faster_rcnn_batchnorm_update\n",
      "INFO:tensorflow:time(__main__.ModelBuilderTF2Test.test_invalid_faster_rcnn_batchnorm_update): 0.0s\n",
      "I0120 09:08:43.344343   320 test_util.py:2462] time(__main__.ModelBuilderTF2Test.test_invalid_faster_rcnn_batchnorm_update): 0.0s\n",
      "[       OK ] ModelBuilderTF2Test.test_invalid_faster_rcnn_batchnorm_update\n",
      "[ RUN      ] ModelBuilderTF2Test.test_invalid_first_stage_nms_iou_threshold\n",
      "INFO:tensorflow:time(__main__.ModelBuilderTF2Test.test_invalid_first_stage_nms_iou_threshold): 0.0s\n",
      "I0120 09:08:43.345510   320 test_util.py:2462] time(__main__.ModelBuilderTF2Test.test_invalid_first_stage_nms_iou_threshold): 0.0s\n",
      "[       OK ] ModelBuilderTF2Test.test_invalid_first_stage_nms_iou_threshold\n",
      "[ RUN      ] ModelBuilderTF2Test.test_invalid_model_config_proto\n",
      "INFO:tensorflow:time(__main__.ModelBuilderTF2Test.test_invalid_model_config_proto): 0.0s\n",
      "I0120 09:08:43.345510   320 test_util.py:2462] time(__main__.ModelBuilderTF2Test.test_invalid_model_config_proto): 0.0s\n",
      "[       OK ] ModelBuilderTF2Test.test_invalid_model_config_proto\n",
      "[ RUN      ] ModelBuilderTF2Test.test_invalid_second_stage_batch_size\n",
      "INFO:tensorflow:time(__main__.ModelBuilderTF2Test.test_invalid_second_stage_batch_size): 0.0s\n",
      "I0120 09:08:43.346620   320 test_util.py:2462] time(__main__.ModelBuilderTF2Test.test_invalid_second_stage_batch_size): 0.0s\n",
      "[       OK ] ModelBuilderTF2Test.test_invalid_second_stage_batch_size\n",
      "[ RUN      ] ModelBuilderTF2Test.test_session\n",
      "[  SKIPPED ] ModelBuilderTF2Test.test_session\n",
      "[ RUN      ] ModelBuilderTF2Test.test_unknown_faster_rcnn_feature_extractor\n",
      "INFO:tensorflow:time(__main__.ModelBuilderTF2Test.test_unknown_faster_rcnn_feature_extractor): 0.0s\n",
      "I0120 09:08:43.346620   320 test_util.py:2462] time(__main__.ModelBuilderTF2Test.test_unknown_faster_rcnn_feature_extractor): 0.0s\n",
      "[       OK ] ModelBuilderTF2Test.test_unknown_faster_rcnn_feature_extractor\n",
      "[ RUN      ] ModelBuilderTF2Test.test_unknown_meta_architecture\n",
      "INFO:tensorflow:time(__main__.ModelBuilderTF2Test.test_unknown_meta_architecture): 0.0s\n",
      "I0120 09:08:43.347819   320 test_util.py:2462] time(__main__.ModelBuilderTF2Test.test_unknown_meta_architecture): 0.0s\n",
      "[       OK ] ModelBuilderTF2Test.test_unknown_meta_architecture\n",
      "[ RUN      ] ModelBuilderTF2Test.test_unknown_ssd_feature_extractor\n",
      "INFO:tensorflow:time(__main__.ModelBuilderTF2Test.test_unknown_ssd_feature_extractor): 0.0s\n",
      "I0120 09:08:43.347819   320 test_util.py:2462] time(__main__.ModelBuilderTF2Test.test_unknown_ssd_feature_extractor): 0.0s\n",
      "[       OK ] ModelBuilderTF2Test.test_unknown_ssd_feature_extractor\n",
      "----------------------------------------------------------------------\n",
      "Ran 24 tests in 28.723s\n",
      "\n",
      "OK (skipped=1)\n"
     ]
    }
   ],
   "source": [
    "VERIFICATION_SCRIPT = os.path.join(paths['APIMODEL_PATH'], 'research', 'object_detection', 'builders', 'model_builder_tf2_test.py')\n",
    "# Verify Installation\n",
    "!python {VERIFICATION_SCRIPT}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: tensorflow in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (2.13.0)\n",
      "Requirement already satisfied: tensorflow-intel==2.13.0 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tensorflow) (2.13.0)\n",
      "Requirement already satisfied: absl-py>=1.0.0 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\absl_py-2.0.0-py3.11.egg (from tensorflow-intel==2.13.0->tensorflow) (2.0.0)\n",
      "Requirement already satisfied: astunparse>=1.6.0 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tensorflow-intel==2.13.0->tensorflow) (1.6.3)\n",
      "Requirement already satisfied: flatbuffers>=23.1.21 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tensorflow-intel==2.13.0->tensorflow) (23.5.26)\n",
      "Requirement already satisfied: gast<=0.4.0,>=0.2.1 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tensorflow-intel==2.13.0->tensorflow) (0.4.0)\n",
      "Requirement already satisfied: google-pasta>=0.1.1 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tensorflow-intel==2.13.0->tensorflow) (0.2.0)\n",
      "Requirement already satisfied: h5py>=2.9.0 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\h5py-3.10.0-py3.11-win-amd64.egg (from tensorflow-intel==2.13.0->tensorflow) (3.10.0)\n",
      "Requirement already satisfied: libclang>=13.0.0 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tensorflow-intel==2.13.0->tensorflow) (16.0.6)\n",
      "Requirement already satisfied: numpy<=1.24.3,>=1.22 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tensorflow-intel==2.13.0->tensorflow) (1.24.0)\n",
      "Requirement already satisfied: opt-einsum>=2.3.2 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tensorflow-intel==2.13.0->tensorflow) (3.3.0)\n",
      "Requirement already satisfied: packaging in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tensorflow-intel==2.13.0->tensorflow) (23.2)\n",
      "Requirement already satisfied: protobuf!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0dev,>=3.20.3 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tensorflow-intel==2.13.0->tensorflow) (4.25.2)\n",
      "Requirement already satisfied: setuptools in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tensorflow-intel==2.13.0->tensorflow) (65.5.0)\n",
      "Requirement already satisfied: six>=1.12.0 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tensorflow-intel==2.13.0->tensorflow) (1.16.0)\n",
      "Requirement already satisfied: termcolor>=1.1.0 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tensorflow-intel==2.13.0->tensorflow) (2.4.0)\n",
      "Requirement already satisfied: typing-extensions<4.6.0,>=3.6.6 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tensorflow-intel==2.13.0->tensorflow) (4.5.0)\n",
      "Requirement already satisfied: wrapt>=1.11.0 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tensorflow-intel==2.13.0->tensorflow) (1.14.1)\n",
      "Requirement already satisfied: grpcio<2.0,>=1.24.3 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tensorflow-intel==2.13.0->tensorflow) (1.60.0)\n",
      "Requirement already satisfied: tensorboard<2.14,>=2.13 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tensorflow-intel==2.13.0->tensorflow) (2.13.0)\n",
      "Requirement already satisfied: tensorflow-estimator<2.14,>=2.13.0 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tensorflow-intel==2.13.0->tensorflow) (2.13.0)\n",
      "Requirement already satisfied: keras<2.14,>=2.13.1 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tensorflow-intel==2.13.0->tensorflow) (2.13.1)\n",
      "Requirement already satisfied: tensorflow-io-gcs-filesystem>=0.23.1 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io_gcs_filesystem-0.31.0-py3.11-win-amd64.egg (from tensorflow-intel==2.13.0->tensorflow) (0.31.0)\n",
      "Requirement already satisfied: wheel<1.0,>=0.23.0 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from astunparse>=1.6.0->tensorflow-intel==2.13.0->tensorflow) (0.42.0)\n",
      "Requirement already satisfied: google-auth<3,>=1.6.3 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tensorboard<2.14,>=2.13->tensorflow-intel==2.13.0->tensorflow) (2.26.2)\n",
      "Requirement already satisfied: google-auth-oauthlib<1.1,>=0.5 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tensorboard<2.14,>=2.13->tensorflow-intel==2.13.0->tensorflow) (1.0.0)\n",
      "Requirement already satisfied: markdown>=2.6.8 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tensorboard<2.14,>=2.13->tensorflow-intel==2.13.0->tensorflow) (3.5.2)\n",
      "Requirement already satisfied: requests<3,>=2.21.0 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tensorboard<2.14,>=2.13->tensorflow-intel==2.13.0->tensorflow) (2.31.0)\n",
      "Requirement already satisfied: tensorboard-data-server<0.8.0,>=0.7.0 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tensorboard<2.14,>=2.13->tensorflow-intel==2.13.0->tensorflow) (0.7.2)\n",
      "Requirement already satisfied: werkzeug>=1.0.1 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tensorboard<2.14,>=2.13->tensorflow-intel==2.13.0->tensorflow) (3.0.1)\n",
      "Requirement already satisfied: cachetools<6.0,>=2.0.0 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from google-auth<3,>=1.6.3->tensorboard<2.14,>=2.13->tensorflow-intel==2.13.0->tensorflow) (5.3.2)\n",
      "Requirement already satisfied: pyasn1-modules>=0.2.1 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from google-auth<3,>=1.6.3->tensorboard<2.14,>=2.13->tensorflow-intel==2.13.0->tensorflow) (0.3.0)\n",
      "Requirement already satisfied: rsa<5,>=3.1.4 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from google-auth<3,>=1.6.3->tensorboard<2.14,>=2.13->tensorflow-intel==2.13.0->tensorflow) (4.9)\n",
      "Requirement already satisfied: requests-oauthlib>=0.7.0 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from google-auth-oauthlib<1.1,>=0.5->tensorboard<2.14,>=2.13->tensorflow-intel==2.13.0->tensorflow) (1.3.1)\n",
      "Requirement already satisfied: charset-normalizer<4,>=2 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from requests<3,>=2.21.0->tensorboard<2.14,>=2.13->tensorflow-intel==2.13.0->tensorflow) (3.3.2)\n",
      "Requirement already satisfied: idna<4,>=2.5 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from requests<3,>=2.21.0->tensorboard<2.14,>=2.13->tensorflow-intel==2.13.0->tensorflow) (3.6)\n",
      "Requirement already satisfied: urllib3<3,>=1.21.1 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from requests<3,>=2.21.0->tensorboard<2.14,>=2.13->tensorflow-intel==2.13.0->tensorflow) (2.1.0)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from requests<3,>=2.21.0->tensorboard<2.14,>=2.13->tensorflow-intel==2.13.0->tensorflow) (2023.11.17)\n",
      "Requirement already satisfied: MarkupSafe>=2.1.1 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from werkzeug>=1.0.1->tensorboard<2.14,>=2.13->tensorflow-intel==2.13.0->tensorflow) (2.1.3)\n",
      "Requirement already satisfied: pyasn1<0.6.0,>=0.4.6 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from pyasn1-modules>=0.2.1->google-auth<3,>=1.6.3->tensorboard<2.14,>=2.13->tensorflow-intel==2.13.0->tensorflow) (0.5.1)\n",
      "Requirement already satisfied: oauthlib>=3.0.0 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from requests-oauthlib>=0.7.0->google-auth-oauthlib<1.1,>=0.5->tensorboard<2.14,>=2.13->tensorflow-intel==2.13.0->tensorflow) (3.2.2)\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\absl_py-2.0.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\apache_beam-2.53.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\avro_python3-1.10.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\contextlib2-21.6.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\cython-3.0.8-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\dm_tree-0.1.8-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\h5py-3.10.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lvis-0.5.3-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lxml-5.1.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\namex-0.0.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\object_detection-0.1-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pandas-2.2.0rc0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pillow-10.2.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\portalocker-2.8.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pycocotools-2.0.7-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pyparsing-2.4.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\regex-2023.12.25-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\rich-13.7.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\sacrebleu-2.2.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\scipy-1.12.0rc2-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tabulate-0.9.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io_gcs_filesystem-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_models_official-2.15.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_slim-1.1.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n"
     ]
    }
   ],
   "source": [
    "pip install tensorflow"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: numpy==1.24.0 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (1.24.0)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\absl_py-2.0.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\apache_beam-2.53.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\avro_python3-1.10.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\contextlib2-21.6.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\cython-3.0.8-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\dm_tree-0.1.8-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\h5py-3.10.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lvis-0.5.3-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lxml-5.1.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\namex-0.0.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\object_detection-0.1-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pandas-2.2.0rc0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pillow-10.2.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\portalocker-2.8.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pycocotools-2.0.7-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pyparsing-2.4.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\regex-2023.12.25-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\rich-13.7.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\sacrebleu-2.2.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\scipy-1.12.0rc2-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tabulate-0.9.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io_gcs_filesystem-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_models_official-2.15.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_slim-1.1.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n"
     ]
    }
   ],
   "source": [
    "!pip install numpy==1.24.0\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found existing installation: protobuf 4.25.2\n",
      "Uninstalling protobuf-4.25.2:\n",
      "  Successfully uninstalled protobuf-4.25.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING: Ignoring invalid distribution ~atplotlib (C:\\Users\\Hasitha\\Desktop\\ANPR\\anprsys\\Lib\\site-packages)\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\absl_py-2.0.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\apache_beam-2.53.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\avro_python3-1.10.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\contextlib2-21.6.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\cython-3.0.8-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\dm_tree-0.1.8-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\h5py-3.10.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lvis-0.5.3-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lxml-5.1.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\namex-0.0.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\object_detection-0.1-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pandas-2.2.0rc0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pillow-10.2.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\portalocker-2.8.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pycocotools-2.0.7-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pyparsing-2.4.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\regex-2023.12.25-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\rich-13.7.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\sacrebleu-2.2.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\scipy-1.12.0rc2-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tabulate-0.9.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io_gcs_filesystem-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_models_official-2.15.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_slim-1.1.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "WARNING: Skipping matplotlib as it is not installed.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting protobuf\n",
      "  Using cached protobuf-4.25.2-cp310-abi3-win_amd64.whl.metadata (541 bytes)\n",
      "Collecting matplotlib==3.2\n",
      "  Using cached matplotlib-3.2.0-cp311-cp311-win_amd64.whl\n",
      "Requirement already satisfied: cycler>=0.10 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from matplotlib==3.2) (0.12.1)\n",
      "Requirement already satisfied: kiwisolver>=1.0.1 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from matplotlib==3.2) (1.4.5)\n",
      "Requirement already satisfied: numpy>=1.11 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from matplotlib==3.2) (1.24.0)\n",
      "Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.1 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pyparsing-2.4.7-py3.11.egg (from matplotlib==3.2) (2.4.7)\n",
      "Requirement already satisfied: python-dateutil>=2.1 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from matplotlib==3.2) (2.8.2)\n",
      "Requirement already satisfied: six>=1.5 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from python-dateutil>=2.1->matplotlib==3.2) (1.16.0)\n",
      "Using cached protobuf-4.25.2-cp310-abi3-win_amd64.whl (413 kB)\n",
      "Installing collected packages: protobuf, matplotlib\n",
      "Successfully installed matplotlib-3.2.0 protobuf-4.25.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING: Ignoring invalid distribution ~atplotlib (C:\\Users\\Hasitha\\Desktop\\ANPR\\anprsys\\Lib\\site-packages)\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\absl_py-2.0.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\apache_beam-2.53.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\avro_python3-1.10.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\contextlib2-21.6.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\cython-3.0.8-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\dm_tree-0.1.8-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\h5py-3.10.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lvis-0.5.3-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lxml-5.1.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\namex-0.0.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\object_detection-0.1-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pandas-2.2.0rc0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pillow-10.2.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\portalocker-2.8.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pycocotools-2.0.7-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pyparsing-2.4.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\regex-2023.12.25-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\rich-13.7.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\sacrebleu-2.2.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\scipy-1.12.0rc2-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tabulate-0.9.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io_gcs_filesystem-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_models_official-2.15.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_slim-1.1.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "WARNING: Ignoring invalid distribution ~atplotlib (C:\\Users\\Hasitha\\Desktop\\ANPR\\anprsys\\Lib\\site-packages)\n",
      "ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n",
      "apache-beam 2.53.0 requires cloudpickle~=2.2.1, which is not installed.\n",
      "apache-beam 2.53.0 requires crcmod<2.0,>=1.7, which is not installed.\n",
      "apache-beam 2.53.0 requires dill<0.3.2,>=0.3.1.1, which is not installed.\n",
      "apache-beam 2.53.0 requires fastavro<2,>=0.23.6, which is not installed.\n",
      "apache-beam 2.53.0 requires fasteners<1.0,>=0.3, which is not installed.\n",
      "apache-beam 2.53.0 requires hdfs<3.0.0,>=2.1.0, which is not installed.\n",
      "apache-beam 2.53.0 requires httplib2<0.23.0,>=0.8, which is not installed.\n",
      "apache-beam 2.53.0 requires js2py<1,>=0.74, which is not installed.\n",
      "apache-beam 2.53.0 requires jsonpickle<4.0.0,>=3.0.0, which is not installed.\n",
      "apache-beam 2.53.0 requires objsize<0.7.0,>=0.6.1, which is not installed.\n",
      "apache-beam 2.53.0 requires orjson<4,>=3.9.7, which is not installed.\n",
      "apache-beam 2.53.0 requires proto-plus<2,>=1.7.1, which is not installed.\n",
      "apache-beam 2.53.0 requires pyarrow<12.0.0,>=3.0.0, which is not installed.\n",
      "apache-beam 2.53.0 requires pyarrow-hotfix<1, which is not installed.\n",
      "apache-beam 2.53.0 requires pydot<2,>=1.2.0, which is not installed.\n",
      "apache-beam 2.53.0 requires pymongo<5.0.0,>=3.8.0, which is not installed.\n",
      "apache-beam 2.53.0 requires zstandard<1,>=0.18.0, which is not installed.\n",
      "tf-models-official 2.15.0 requires google-api-python-client>=1.6.7, which is not installed.\n",
      "tf-models-official 2.15.0 requires immutabledict, which is not installed.\n",
      "tf-models-official 2.15.0 requires kaggle>=1.3.9, which is not installed.\n",
      "tf-models-official 2.15.0 requires oauth2client, which is not installed.\n",
      "tf-models-official 2.15.0 requires opencv-python-headless, which is not installed.\n",
      "tf-models-official 2.15.0 requires py-cpuinfo>=3.3.0, which is not installed.\n",
      "tf-models-official 2.15.0 requires sentencepiece, which is not installed.\n",
      "tf-models-official 2.15.0 requires seqeval, which is not installed.\n",
      "tf-models-official 2.15.0 requires tensorflow-datasets, which is not installed.\n",
      "tf-models-official 2.15.0 requires tensorflow-hub>=0.6.0, which is not installed.\n",
      "tf-models-official 2.15.0 requires tensorflow-model-optimization>=0.4.1, which is not installed.\n",
      "tf-models-official 2.15.0 requires tensorflow-text~=2.15.0, which is not installed.\n",
      "tf-models-official 2.15.0 requires tensorflow~=2.15.0, but you have tensorflow 2.13.0 which is incompatible.\n"
     ]
    }
   ],
   "source": [
    "!pip uninstall protobuf matplotlib -y\n",
    "!pip install protobuf matplotlib==3.2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting protobuf==3.20.3\n",
      "  Using cached protobuf-3.20.3-py2.py3-none-any.whl (162 kB)\n",
      "Installing collected packages: protobuf\n",
      "  Attempting uninstall: protobuf\n",
      "    Found existing installation: protobuf 4.25.2\n",
      "    Uninstalling protobuf-4.25.2:\n",
      "      Successfully uninstalled protobuf-4.25.2\n",
      "Successfully installed protobuf-3.20.3\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\absl_py-2.0.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\apache_beam-2.53.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\avro_python3-1.10.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\contextlib2-21.6.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\cython-3.0.8-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\dm_tree-0.1.8-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\h5py-3.10.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lvis-0.5.3-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lxml-5.1.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\namex-0.0.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\object_detection-0.1-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pandas-2.2.0rc0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pillow-10.2.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\portalocker-2.8.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pycocotools-2.0.7-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pyparsing-2.4.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\regex-2023.12.25-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\rich-13.7.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\sacrebleu-2.2.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\scipy-1.12.0rc2-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tabulate-0.9.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io_gcs_filesystem-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_models_official-2.15.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_slim-1.1.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n",
      "apache-beam 2.53.0 requires cloudpickle~=2.2.1, which is not installed.\n",
      "apache-beam 2.53.0 requires crcmod<2.0,>=1.7, which is not installed.\n",
      "apache-beam 2.53.0 requires dill<0.3.2,>=0.3.1.1, which is not installed.\n",
      "apache-beam 2.53.0 requires fastavro<2,>=0.23.6, which is not installed.\n",
      "apache-beam 2.53.0 requires fasteners<1.0,>=0.3, which is not installed.\n",
      "apache-beam 2.53.0 requires hdfs<3.0.0,>=2.1.0, which is not installed.\n",
      "apache-beam 2.53.0 requires httplib2<0.23.0,>=0.8, which is not installed.\n",
      "apache-beam 2.53.0 requires js2py<1,>=0.74, which is not installed.\n",
      "apache-beam 2.53.0 requires jsonpickle<4.0.0,>=3.0.0, which is not installed.\n",
      "apache-beam 2.53.0 requires objsize<0.7.0,>=0.6.1, which is not installed.\n",
      "apache-beam 2.53.0 requires orjson<4,>=3.9.7, which is not installed.\n",
      "apache-beam 2.53.0 requires proto-plus<2,>=1.7.1, which is not installed.\n",
      "apache-beam 2.53.0 requires pyarrow<12.0.0,>=3.0.0, which is not installed.\n",
      "apache-beam 2.53.0 requires pyarrow-hotfix<1, which is not installed.\n",
      "apache-beam 2.53.0 requires pydot<2,>=1.2.0, which is not installed.\n",
      "apache-beam 2.53.0 requires pymongo<5.0.0,>=3.8.0, which is not installed.\n",
      "apache-beam 2.53.0 requires zstandard<1,>=0.18.0, which is not installed.\n"
     ]
    }
   ],
   "source": [
    "!pip install protobuf==3.20.3\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting cloudpickle\n",
      "  Using cached cloudpickle-3.0.0-py3-none-any.whl.metadata (7.0 kB)\n",
      "Collecting crcmod\n",
      "  Using cached crcmod-1.7.tar.gz (89 kB)\n",
      "  Preparing metadata (setup.py): started\n",
      "  Preparing metadata (setup.py): finished with status 'done'\n",
      "Collecting dill\n",
      "  Using cached dill-0.3.7-py3-none-any.whl.metadata (9.9 kB)\n",
      "Collecting fastavro\n",
      "  Using cached fastavro-1.9.3-cp311-cp311-win_amd64.whl.metadata (5.7 kB)\n",
      "Collecting fasteners\n",
      "  Using cached fasteners-0.19-py3-none-any.whl.metadata (4.9 kB)\n",
      "Collecting hdfs\n",
      "  Using cached hdfs-2.7.3.tar.gz (43 kB)\n",
      "  Preparing metadata (setup.py): started\n",
      "  Preparing metadata (setup.py): finished with status 'done'\n",
      "Collecting httplib2\n",
      "  Using cached httplib2-0.22.0-py3-none-any.whl (96 kB)\n",
      "Collecting js2py\n",
      "  Using cached Js2Py-0.74-py3-none-any.whl (1.0 MB)\n",
      "Collecting jsonpickle\n",
      "  Using cached jsonpickle-3.0.2-py3-none-any.whl.metadata (7.5 kB)\n",
      "Collecting objsize\n",
      "  Using cached objsize-0.7.0-py3-none-any.whl.metadata (12 kB)\n",
      "Collecting orjson\n",
      "  Downloading orjson-3.9.12-cp311-none-win_amd64.whl.metadata (50 kB)\n",
      "     ---------------------------------------- 0.0/50.6 kB ? eta -:--:--\n",
      "     --------------- ---------------------- 20.5/50.6 kB 640.0 kB/s eta 0:00:01\n",
      "     ----------------------- -------------- 30.7/50.6 kB 435.7 kB/s eta 0:00:01\n",
      "     ----------------------- -------------- 30.7/50.6 kB 435.7 kB/s eta 0:00:01\n",
      "     ----------------------- -------------- 30.7/50.6 kB 435.7 kB/s eta 0:00:01\n",
      "     -------------------------------------- 50.6/50.6 kB 184.3 kB/s eta 0:00:00\n",
      "Collecting proto-plus\n",
      "  Using cached proto_plus-1.23.0-py3-none-any.whl.metadata (2.2 kB)\n",
      "Collecting pyarrow\n",
      "  Using cached pyarrow-14.0.2-cp311-cp311-win_amd64.whl.metadata (3.1 kB)\n",
      "Collecting pydot\n",
      "  Using cached pydot-2.0.0-py3-none-any.whl.metadata (9.6 kB)\n",
      "Collecting pymongo\n",
      "  Using cached pymongo-4.6.1-cp311-cp311-win_amd64.whl.metadata (22 kB)\n",
      "Requirement already satisfied: pytz in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (2023.3.post1)\n",
      "Collecting zstandard\n",
      "  Using cached zstandard-0.22.0-cp311-cp311-win_amd64.whl.metadata (3.0 kB)\n",
      "Requirement already satisfied: gin-config in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (0.5.0)\n",
      "Collecting google-api-python-client\n",
      "  Downloading google_api_python_client-2.114.0-py2.py3-none-any.whl.metadata (6.6 kB)\n",
      "Collecting immutabledict\n",
      "  Using cached immutabledict-4.1.0-py3-none-any.whl.metadata (3.2 kB)\n",
      "Collecting kaggle\n",
      "  Using cached kaggle-1.6.3.tar.gz (84 kB)\n",
      "  Preparing metadata (setup.py): started\n",
      "  Preparing metadata (setup.py): finished with status 'done'\n",
      "Collecting oauth2client\n",
      "  Using cached oauth2client-4.1.3-py2.py3-none-any.whl (98 kB)\n",
      "Collecting opencv-python-headless\n",
      "  Using cached opencv_python_headless-4.9.0.80-cp37-abi3-win_amd64.whl.metadata (20 kB)\n",
      "Collecting py-cpuinfo\n",
      "  Using cached py_cpuinfo-9.0.0-py3-none-any.whl (22 kB)\n",
      "Collecting sentencepiece\n",
      "  Using cached sentencepiece-0.1.99-cp311-cp311-win_amd64.whl (977 kB)\n",
      "Collecting seqeval\n",
      "  Using cached seqeval-1.2.2.tar.gz (43 kB)\n",
      "  Preparing metadata (setup.py): started\n",
      "  Preparing metadata (setup.py): finished with status 'done'\n",
      "Collecting tensorflow-datasets\n",
      "  Using cached tensorflow_datasets-4.9.4-py3-none-any.whl.metadata (9.2 kB)\n",
      "Collecting tensorflow-hub\n",
      "  Using cached tensorflow_hub-0.15.0-py2.py3-none-any.whl.metadata (1.3 kB)\n",
      "Collecting tensorflow-model-optimization\n",
      "  Using cached tensorflow_model_optimization-0.7.5-py2.py3-none-any.whl.metadata (914 bytes)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\absl_py-2.0.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\apache_beam-2.53.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\avro_python3-1.10.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\contextlib2-21.6.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\cython-3.0.8-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\dm_tree-0.1.8-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\h5py-3.10.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lvis-0.5.3-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lxml-5.1.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\namex-0.0.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\object_detection-0.1-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pandas-2.2.0rc0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pillow-10.2.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\portalocker-2.8.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pycocotools-2.0.7-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pyparsing-2.4.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\regex-2023.12.25-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\rich-13.7.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\sacrebleu-2.2.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\scipy-1.12.0rc2-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tabulate-0.9.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io_gcs_filesystem-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_models_official-2.15.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_slim-1.1.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "ERROR: Could not find a version that satisfies the requirement tensorflow-text (from versions: none)\n",
      "ERROR: No matching distribution found for tensorflow-text\n"
     ]
    }
   ],
   "source": [
    "!pip install cloudpickle crcmod dill fastavro fasteners hdfs httplib2 js2py jsonpickle objsize orjson proto-plus pyarrow pydot pymongo pytz zstandard gin-config google-api-python-client immutabledict kaggle oauth2client opencv-python-headless py-cpuinfo sentencepiece seqeval tensorflow-datasets tensorflow-hub tensorflow-model-optimization tensorflow-text\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: tf-models-official in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_models_official-2.15.0-py3.11.egg (2.15.0)\n",
      "Requirement already satisfied: Cython in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\cython-3.0.8-py3.11.egg (from tf-models-official) (3.0.8)\n",
      "Requirement already satisfied: Pillow in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pillow-10.2.0-py3.11-win-amd64.egg (from tf-models-official) (10.2.0)\n",
      "Requirement already satisfied: gin-config in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tf-models-official) (0.5.0)\n",
      "Collecting google-api-python-client>=1.6.7 (from tf-models-official)\n",
      "  Using cached google_api_python_client-2.114.0-py2.py3-none-any.whl.metadata (6.6 kB)\n",
      "Collecting immutabledict (from tf-models-official)\n",
      "  Using cached immutabledict-4.1.0-py3-none-any.whl.metadata (3.2 kB)\n",
      "Collecting kaggle>=1.3.9 (from tf-models-official)\n",
      "  Using cached kaggle-1.6.3.tar.gz (84 kB)\n",
      "  Preparing metadata (setup.py): started\n",
      "  Preparing metadata (setup.py): finished with status 'done'\n",
      "Requirement already satisfied: matplotlib in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tf-models-official) (3.2.0)\n",
      "Requirement already satisfied: numpy>=1.20 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tf-models-official) (1.24.0)\n",
      "Collecting oauth2client (from tf-models-official)\n",
      "  Using cached oauth2client-4.1.3-py2.py3-none-any.whl (98 kB)\n",
      "Collecting opencv-python-headless (from tf-models-official)\n",
      "  Using cached opencv_python_headless-4.9.0.80-cp37-abi3-win_amd64.whl.metadata (20 kB)\n",
      "Requirement already satisfied: pandas>=0.22.0 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pandas-2.2.0rc0-py3.11-win-amd64.egg (from tf-models-official) (2.2.0rc0)\n",
      "Requirement already satisfied: psutil>=5.4.3 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tf-models-official) (5.9.7)\n",
      "Collecting py-cpuinfo>=3.3.0 (from tf-models-official)\n",
      "  Using cached py_cpuinfo-9.0.0-py3-none-any.whl (22 kB)\n",
      "Requirement already satisfied: pycocotools in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pycocotools-2.0.7-py3.11-win-amd64.egg (from tf-models-official) (2.0.7)\n",
      "Requirement already satisfied: pyyaml>=6.0.0 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tf-models-official) (6.0.1)\n",
      "Requirement already satisfied: sacrebleu in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\sacrebleu-2.2.0-py3.11.egg (from tf-models-official) (2.2.0)\n",
      "Requirement already satisfied: scipy>=0.19.1 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\scipy-1.12.0rc2-py3.11-win-amd64.egg (from tf-models-official) (1.12.0rc2)\n",
      "Collecting sentencepiece (from tf-models-official)\n",
      "  Using cached sentencepiece-0.1.99-cp311-cp311-win_amd64.whl (977 kB)\n",
      "Collecting seqeval (from tf-models-official)\n",
      "  Using cached seqeval-1.2.2.tar.gz (43 kB)\n",
      "  Preparing metadata (setup.py): started\n",
      "  Preparing metadata (setup.py): finished with status 'done'\n",
      "Requirement already satisfied: six in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from tf-models-official) (1.16.0)\n",
      "Collecting tensorflow-datasets (from tf-models-official)\n",
      "  Using cached tensorflow_datasets-4.9.4-py3-none-any.whl.metadata (9.2 kB)\n",
      "Collecting tensorflow-hub>=0.6.0 (from tf-models-official)\n",
      "  Using cached tensorflow_hub-0.15.0-py2.py3-none-any.whl.metadata (1.3 kB)\n",
      "Collecting tensorflow-model-optimization>=0.4.1 (from tf-models-official)\n",
      "  Using cached tensorflow_model_optimization-0.7.5-py2.py3-none-any.whl.metadata (914 bytes)\n",
      "INFO: pip is looking at multiple versions of tf-models-official to determine which version is compatible with other requirements. This could take a while.\n",
      "Collecting tf-models-official\n",
      "  Using cached tf_models_official-2.15.0-py2.py3-none-any.whl.metadata (1.4 kB)\n",
      "  Using cached tf_models_official-2.14.2-py2.py3-none-any.whl.metadata (1.4 kB)\n",
      "  Using cached tf_models_official-2.14.1-py2.py3-none-any.whl.metadata (1.4 kB)\n",
      "  Using cached tf_models_official-2.14.0-py2.py3-none-any.whl.metadata (1.4 kB)\n",
      "INFO: pip is still looking at multiple versions of tf-models-official to determine which version is compatible with other requirements. This could take a while.\n",
      "  Using cached tf_models_official-2.13.2-py2.py3-none-any.whl.metadata (1.4 kB)\n",
      "  Using cached tf_models_official-2.13.1-py2.py3-none-any.whl.metadata (1.4 kB)\n",
      "Collecting pyyaml<5.4.0,>=5.1 (from tf-models-official)\n",
      "  Using cached PyYAML-5.3.1.tar.gz (269 kB)\n",
      "  Preparing metadata (setup.py): started\n",
      "  Preparing metadata (setup.py): finished with status 'error'\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\absl_py-2.0.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\apache_beam-2.53.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\avro_python3-1.10.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\contextlib2-21.6.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\cython-3.0.8-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\dm_tree-0.1.8-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\h5py-3.10.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lvis-0.5.3-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lxml-5.1.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\namex-0.0.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\object_detection-0.1-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pandas-2.2.0rc0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pillow-10.2.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\portalocker-2.8.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pycocotools-2.0.7-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pyparsing-2.4.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\regex-2023.12.25-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\rich-13.7.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\sacrebleu-2.2.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\scipy-1.12.0rc2-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tabulate-0.9.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io_gcs_filesystem-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_models_official-2.15.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_slim-1.1.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "  error: subprocess-exited-with-error\n",
      "  \n",
      "  python setup.py egg_info did not run successfully.\n",
      "  exit code: 1\n",
      "  \n",
      "  [45 lines of output]\n",
      "  C:\\Users\\Hasitha\\Desktop\\ANPR\\anprsys\\Lib\\site-packages\\setuptools\\config\\setupcfg.py:508: SetuptoolsDeprecationWarning: The license_file parameter is deprecated, use license_files instead.\n",
      "    warnings.warn(msg, warning_class)\n",
      "  running egg_info\n",
      "  creating C:\\Users\\Hasitha\\AppData\\Local\\Temp\\pip-pip-egg-info-e9zpk4dk\\PyYAML.egg-info\n",
      "  writing C:\\Users\\Hasitha\\AppData\\Local\\Temp\\pip-pip-egg-info-e9zpk4dk\\PyYAML.egg-info\\PKG-INFO\n",
      "  writing dependency_links to C:\\Users\\Hasitha\\AppData\\Local\\Temp\\pip-pip-egg-info-e9zpk4dk\\PyYAML.egg-info\\dependency_links.txt\n",
      "  writing top-level names to C:\\Users\\Hasitha\\AppData\\Local\\Temp\\pip-pip-egg-info-e9zpk4dk\\PyYAML.egg-info\\top_level.txt\n",
      "  writing manifest file 'C:\\Users\\Hasitha\\AppData\\Local\\Temp\\pip-pip-egg-info-e9zpk4dk\\PyYAML.egg-info\\SOURCES.txt'\n",
      "  Traceback (most recent call last):\n",
      "    File \"<string>\", line 2, in <module>\n",
      "    File \"<pip-setuptools-caller>\", line 34, in <module>\n",
      "    File \"C:\\Users\\Hasitha\\AppData\\Local\\Temp\\pip-install-mfj9xxky\\pyyaml_e12f884b26f747c7a794df137eb4dcff\\setup.py\", line 291, in <module>\n",
      "      setup(\n",
      "    File \"C:\\Users\\Hasitha\\Desktop\\ANPR\\anprsys\\Lib\\site-packages\\setuptools\\_distutils\\core.py\", line 185, in setup\n",
      "      return run_commands(dist)\n",
      "             ^^^^^^^^^^^^^^^^^^\n",
      "    File \"C:\\Users\\Hasitha\\Desktop\\ANPR\\anprsys\\Lib\\site-packages\\setuptools\\_distutils\\core.py\", line 201, in run_commands\n",
      "      dist.run_commands()\n",
      "    File \"C:\\Users\\Hasitha\\Desktop\\ANPR\\anprsys\\Lib\\site-packages\\setuptools\\_distutils\\dist.py\", line 968, in run_commands\n",
      "      self.run_command(cmd)\n",
      "    File \"C:\\Users\\Hasitha\\Desktop\\ANPR\\anprsys\\Lib\\site-packages\\setuptools\\dist.py\", line 1217, in run_command\n",
      "      super().run_command(command)\n",
      "    File \"C:\\Users\\Hasitha\\Desktop\\ANPR\\anprsys\\Lib\\site-packages\\setuptools\\_distutils\\dist.py\", line 987, in run_command\n",
      "      cmd_obj.run()\n",
      "    File \"C:\\Users\\Hasitha\\Desktop\\ANPR\\anprsys\\Lib\\site-packages\\setuptools\\command\\egg_info.py\", line 308, in run\n",
      "      self.find_sources()\n",
      "    File \"C:\\Users\\Hasitha\\Desktop\\ANPR\\anprsys\\Lib\\site-packages\\setuptools\\command\\egg_info.py\", line 316, in find_sources\n",
      "      mm.run()\n",
      "    File \"C:\\Users\\Hasitha\\Desktop\\ANPR\\anprsys\\Lib\\site-packages\\setuptools\\command\\egg_info.py\", line 560, in run\n",
      "      self.add_defaults()\n",
      "    File \"C:\\Users\\Hasitha\\Desktop\\ANPR\\anprsys\\Lib\\site-packages\\setuptools\\command\\egg_info.py\", line 597, in add_defaults\n",
      "      sdist.add_defaults(self)\n",
      "    File \"C:\\Users\\Hasitha\\Desktop\\ANPR\\anprsys\\Lib\\site-packages\\setuptools\\command\\sdist.py\", line 106, in add_defaults\n",
      "      super().add_defaults()\n",
      "    File \"C:\\Users\\Hasitha\\Desktop\\ANPR\\anprsys\\Lib\\site-packages\\setuptools\\_distutils\\command\\sdist.py\", line 252, in add_defaults\n",
      "      self._add_defaults_ext()\n",
      "    File \"C:\\Users\\Hasitha\\Desktop\\ANPR\\anprsys\\Lib\\site-packages\\setuptools\\_distutils\\command\\sdist.py\", line 337, in _add_defaults_ext\n",
      "      self.filelist.extend(build_ext.get_source_files())\n",
      "                           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n",
      "    File \"C:\\Users\\Hasitha\\AppData\\Local\\Temp\\pip-install-mfj9xxky\\pyyaml_e12f884b26f747c7a794df137eb4dcff\\setup.py\", line 199, in get_source_files\n",
      "      self.cython_sources(ext.sources, ext)\n",
      "      ^^^^^^^^^^^^^^^^^^^\n",
      "    File \"C:\\Users\\Hasitha\\Desktop\\ANPR\\anprsys\\Lib\\site-packages\\setuptools\\_distutils\\cmd.py\", line 105, in __getattr__\n",
      "      raise AttributeError(attr)\n",
      "  AttributeError: cython_sources\n",
      "  [end of output]\n",
      "  \n",
      "  note: This error originates from a subprocess, and is likely not a problem with pip.\n",
      "error: metadata-generation-failed\n",
      "\n",
      "Encountered error while generating package metadata.\n",
      "\n",
      "See above for output.\n",
      "\n",
      "note: This is an issue with the package mentioned above, not pip.\n",
      "hint: See above for details.\n"
     ]
    }
   ],
   "source": [
    "!pip install --upgrade tf-models-official\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "import object_detection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "!pip list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "csofht2npfDE",
    "outputId": "ff5471b2-bed2-43f2-959c-327a706527b6"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "100% [........................................................................] 20515344 / 20515344        1 file(s) moved.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "x ssd_mobilenet_v2_fpnlite_320x320_coco17_tpu-8/\n",
      "x ssd_mobilenet_v2_fpnlite_320x320_coco17_tpu-8/checkpoint/\n",
      "x ssd_mobilenet_v2_fpnlite_320x320_coco17_tpu-8/checkpoint/ckpt-0.data-00000-of-00001\n",
      "x ssd_mobilenet_v2_fpnlite_320x320_coco17_tpu-8/checkpoint/checkpoint\n",
      "x ssd_mobilenet_v2_fpnlite_320x320_coco17_tpu-8/checkpoint/ckpt-0.index\n",
      "x ssd_mobilenet_v2_fpnlite_320x320_coco17_tpu-8/pipeline.config\n",
      "x ssd_mobilenet_v2_fpnlite_320x320_coco17_tpu-8/saved_model/\n",
      "x ssd_mobilenet_v2_fpnlite_320x320_coco17_tpu-8/saved_model/saved_model.pb\n",
      "x ssd_mobilenet_v2_fpnlite_320x320_coco17_tpu-8/saved_model/variables/\n",
      "x ssd_mobilenet_v2_fpnlite_320x320_coco17_tpu-8/saved_model/variables/variables.data-00000-of-00001\n",
      "x ssd_mobilenet_v2_fpnlite_320x320_coco17_tpu-8/saved_model/variables/variables.index\n"
     ]
    }
   ],
   "source": [
    "if os.name =='posix':\n",
    "    !wget {PRETRAINED_MODEL_URL}\n",
    "    !mv {PRETRAINED_MODEL_NAME+'.tar.gz'} {paths['PRETRAINED_MODEL_PATH']}\n",
    "    !cd {paths['PRETRAINED_MODEL_PATH']} && tar -zxvf {PRETRAINED_MODEL_NAME+'.tar.gz'}\n",
    "if os.name == 'nt':\n",
    "    wget.download(PRETRAINED_MODEL_URL)\n",
    "    !move {PRETRAINED_MODEL_NAME+'.tar.gz'} {paths['PRETRAINED_MODEL_PATH']}\n",
    "    !cd {paths['PRETRAINED_MODEL_PATH']} && tar -zxvf {PRETRAINED_MODEL_NAME+'.tar.gz'}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "M5KJTnkfpfDC"
   },
   "source": [
    "# 2. Create Label Map"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "id": "p1BVDWo7pfDC"
   },
   "outputs": [],
   "source": [
    "labels = [{'name':'licence', 'id':1}]\n",
    "\n",
    "with open(files['LABELMAP'], 'w') as f:\n",
    "    for label in labels:\n",
    "        f.write('item { \\n')\n",
    "        f.write('\\tname:\\'{}\\'\\n'.format(label['name']))\n",
    "        f.write('\\tid:{}\\n'.format(label['id']))\n",
    "        f.write('}\\n')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "C88zyVELpfDC"
   },
   "source": [
    "# 3. Create TF records"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "kvf5WccwrFGq",
    "outputId": "49902aeb-0bd7-4298-e1a0-5b4a64eb2064"
   },
   "outputs": [],
   "source": [
    "# OPTIONAL IF RUNNING ON COLAB\n",
    "ARCHIVE_FILES = os.path.join(paths['IMAGE_PATH'], 'archive.tar.gz')\n",
    "if os.path.exists(ARCHIVE_FILES):\n",
    "  !tar -zxvf {ARCHIVE_FILES}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "KWpb_BVUpfDD",
    "outputId": "56ce2a3f-3933-4ee6-8a9d-d5ec65f7d73c"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Cloning into 'Tensorflow\\scripts'...\n"
     ]
    }
   ],
   "source": [
    "if not os.path.exists(files['TF_RECORD_SCRIPT']):\n",
    "    !git clone https://github.com/nicknochnack/GenerateTFRecord {paths['SCRIPTS_PATH']}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting pytz\n",
      "  Using cached pytz-2023.3.post1-py2.py3-none-any.whl.metadata (22 kB)\n",
      "Downloading pytz-2023.3.post1-py2.py3-none-any.whl (502 kB)\n",
      "   ---------------------------------------- 0.0/502.5 kB ? eta -:--:--\n",
      "   -- ------------------------------------- 30.7/502.5 kB 1.4 MB/s eta 0:00:01\n",
      "   -- ------------------------------------- 30.7/502.5 kB 1.4 MB/s eta 0:00:01\n",
      "   ------ -------------------------------- 81.9/502.5 kB 657.6 kB/s eta 0:00:01\n",
      "   ------- ------------------------------- 92.2/502.5 kB 525.1 kB/s eta 0:00:01\n",
      "   ---------- --------------------------- 143.4/502.5 kB 655.8 kB/s eta 0:00:01\n",
      "   -------------- ----------------------- 194.6/502.5 kB 737.3 kB/s eta 0:00:01\n",
      "   ----------------- -------------------- 225.3/502.5 kB 765.3 kB/s eta 0:00:01\n",
      "   ---------------------- --------------- 297.0/502.5 kB 833.5 kB/s eta 0:00:01\n",
      "   -------------------------- ----------- 348.2/502.5 kB 901.1 kB/s eta 0:00:01\n",
      "   ------------------------------ ------- 399.4/502.5 kB 890.4 kB/s eta 0:00:01\n",
      "   ------------------------------------ - 481.3/502.5 kB 973.2 kB/s eta 0:00:01\n",
      "   -------------------------------------- 502.5/502.5 kB 955.8 kB/s eta 0:00:00\n",
      "Installing collected packages: pytz\n",
      "Successfully installed pytz-2023.3.post1\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\absl_py-2.0.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\apache_beam-2.53.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\avro_python3-1.10.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\contextlib2-21.6.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\cython-3.0.8-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\dm_tree-0.1.8-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\h5py-3.10.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lvis-0.5.3-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lxml-5.1.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\namex-0.0.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\object_detection-0.1-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pandas-2.2.0rc0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pillow-10.2.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\portalocker-2.8.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pycocotools-2.0.7-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pyparsing-2.4.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\regex-2023.12.25-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\rich-13.7.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\sacrebleu-2.2.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\scipy-1.12.0rc2-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tabulate-0.9.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow-2.15.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io_gcs_filesystem-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_models_official-2.15.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_slim-1.1.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n",
      "apache-beam 2.53.0 requires cloudpickle~=2.2.1, which is not installed.\n",
      "apache-beam 2.53.0 requires crcmod<2.0,>=1.7, which is not installed.\n",
      "apache-beam 2.53.0 requires dill<0.3.2,>=0.3.1.1, which is not installed.\n",
      "apache-beam 2.53.0 requires fastavro<2,>=0.23.6, which is not installed.\n",
      "apache-beam 2.53.0 requires fasteners<1.0,>=0.3, which is not installed.\n",
      "apache-beam 2.53.0 requires hdfs<3.0.0,>=2.1.0, which is not installed.\n",
      "apache-beam 2.53.0 requires httplib2<0.23.0,>=0.8, which is not installed.\n",
      "apache-beam 2.53.0 requires js2py<1,>=0.74, which is not installed.\n",
      "apache-beam 2.53.0 requires jsonpickle<4.0.0,>=3.0.0, which is not installed.\n",
      "apache-beam 2.53.0 requires objsize<0.7.0,>=0.6.1, which is not installed.\n",
      "apache-beam 2.53.0 requires orjson<4,>=3.9.7, which is not installed.\n",
      "apache-beam 2.53.0 requires proto-plus<2,>=1.7.1, which is not installed.\n",
      "apache-beam 2.53.0 requires pyarrow<12.0.0,>=3.0.0, which is not installed.\n",
      "apache-beam 2.53.0 requires pyarrow-hotfix<1, which is not installed.\n",
      "apache-beam 2.53.0 requires pydot<2,>=1.2.0, which is not installed.\n",
      "apache-beam 2.53.0 requires pymongo<5.0.0,>=3.8.0, which is not installed.\n",
      "apache-beam 2.53.0 requires zstandard<1,>=0.18.0, which is not installed.\n",
      "pandas 2.2.0rc0 requires tzdata>=2022.7, which is not installed.\n",
      "tf-models-official 2.15.0 requires gin-config, which is not installed.\n",
      "tf-models-official 2.15.0 requires google-api-python-client>=1.6.7, which is not installed.\n",
      "tf-models-official 2.15.0 requires immutabledict, which is not installed.\n",
      "tf-models-official 2.15.0 requires kaggle>=1.3.9, which is not installed.\n",
      "tf-models-official 2.15.0 requires oauth2client, which is not installed.\n",
      "tf-models-official 2.15.0 requires opencv-python-headless, which is not installed.\n",
      "tf-models-official 2.15.0 requires py-cpuinfo>=3.3.0, which is not installed.\n",
      "tf-models-official 2.15.0 requires sentencepiece, which is not installed.\n",
      "tf-models-official 2.15.0 requires seqeval, which is not installed.\n",
      "tf-models-official 2.15.0 requires tensorflow-datasets, which is not installed.\n",
      "tf-models-official 2.15.0 requires tensorflow-hub>=0.6.0, which is not installed.\n",
      "tf-models-official 2.15.0 requires tensorflow-model-optimization>=0.4.1, which is not installed.\n",
      "tf-models-official 2.15.0 requires tensorflow-text~=2.15.0, which is not installed.\n"
     ]
    }
   ],
   "source": [
    "!pip install pytz"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "UPFToGZqpfDD",
    "outputId": "0ebb456f-aadc-4a1f-96e6-fbfec1923e1c"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully created the TFRecord file: Tensorflow\\workspace\\annotations\\train.record\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From C:\\Users\\Hasitha\\Desktop\\ANPR\\anprsys\\Lib\\site-packages\\keras\\src\\losses.py:2976: The name tf.losses.sparse_softmax_cross_entropy is deprecated. Please use tf.compat.v1.losses.sparse_softmax_cross_entropy instead.\n",
      "\n",
      "WARNING:tensorflow:From C:\\Users\\Hasitha\\Desktop\\ANPR\\Tensorflow\\scripts\\generate_tfrecord.py:168: The name tf.app.run is deprecated. Please use tf.compat.v1.app.run instead.\n",
      "\n",
      "WARNING:tensorflow:From C:\\Users\\Hasitha\\Desktop\\ANPR\\Tensorflow\\scripts\\generate_tfrecord.py:153: The name tf.python_io.TFRecordWriter is deprecated. Please use tf.io.TFRecordWriter instead.\n",
      "\n",
      "W0115 00:02:25.258640 26888 module_wrapper.py:149] From C:\\Users\\Hasitha\\Desktop\\ANPR\\Tensorflow\\scripts\\generate_tfrecord.py:153: The name tf.python_io.TFRecordWriter is deprecated. Please use tf.io.TFRecordWriter instead.\n",
      "\n",
      "WARNING:tensorflow:From C:\\Users\\Hasitha\\Desktop\\ANPR\\Tensorflow\\scripts\\generate_tfrecord.py:111: The name tf.gfile.GFile is deprecated. Please use tf.io.gfile.GFile instead.\n",
      "\n",
      "W0115 00:02:25.381089 26888 module_wrapper.py:149] From C:\\Users\\Hasitha\\Desktop\\ANPR\\Tensorflow\\scripts\\generate_tfrecord.py:111: The name tf.gfile.GFile is deprecated. Please use tf.io.gfile.GFile instead.\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully created the TFRecord file: Tensorflow\\workspace\\annotations\\test.record\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From C:\\Users\\Hasitha\\Desktop\\ANPR\\anprsys\\Lib\\site-packages\\keras\\src\\losses.py:2976: The name tf.losses.sparse_softmax_cross_entropy is deprecated. Please use tf.compat.v1.losses.sparse_softmax_cross_entropy instead.\n",
      "\n",
      "WARNING:tensorflow:From C:\\Users\\Hasitha\\Desktop\\ANPR\\Tensorflow\\scripts\\generate_tfrecord.py:168: The name tf.app.run is deprecated. Please use tf.compat.v1.app.run instead.\n",
      "\n",
      "WARNING:tensorflow:From C:\\Users\\Hasitha\\Desktop\\ANPR\\Tensorflow\\scripts\\generate_tfrecord.py:153: The name tf.python_io.TFRecordWriter is deprecated. Please use tf.io.TFRecordWriter instead.\n",
      "\n",
      "W0115 00:02:32.424550  6924 module_wrapper.py:149] From C:\\Users\\Hasitha\\Desktop\\ANPR\\Tensorflow\\scripts\\generate_tfrecord.py:153: The name tf.python_io.TFRecordWriter is deprecated. Please use tf.io.TFRecordWriter instead.\n",
      "\n",
      "WARNING:tensorflow:From C:\\Users\\Hasitha\\Desktop\\ANPR\\Tensorflow\\scripts\\generate_tfrecord.py:111: The name tf.gfile.GFile is deprecated. Please use tf.io.gfile.GFile instead.\n",
      "\n",
      "W0115 00:02:32.435957  6924 module_wrapper.py:149] From C:\\Users\\Hasitha\\Desktop\\ANPR\\Tensorflow\\scripts\\generate_tfrecord.py:111: The name tf.gfile.GFile is deprecated. Please use tf.io.gfile.GFile instead.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "!python {files['TF_RECORD_SCRIPT']} -x {os.path.join(paths['IMAGE_PATH'], 'train')} -l {files['LABELMAP']} -o {os.path.join(paths['ANNOTATION_PATH'], 'train.record')} \n",
    "!python {files['TF_RECORD_SCRIPT']} -x {os.path.join(paths['IMAGE_PATH'], 'test')} -l {files['LABELMAP']} -o {os.path.join(paths['ANNOTATION_PATH'], 'test.record')} "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "qT4QU7pLpfDE"
   },
   "source": [
    "# 4. Copy Model Config to Training Folder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "id": "cOjuTFbwpfDF"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "        1 file(s) copied.\n"
     ]
    }
   ],
   "source": [
    "if os.name =='posix':\n",
    "    !cp {os.path.join(paths['PRETRAINED_MODEL_PATH'], PRETRAINED_MODEL_NAME, 'pipeline.config')} {os.path.join(paths['CHECKPOINT_PATH'])}\n",
    "if os.name == 'nt':\n",
    "    !copy {os.path.join(paths['PRETRAINED_MODEL_PATH'], PRETRAINED_MODEL_NAME, 'pipeline.config')} {os.path.join(paths['CHECKPOINT_PATH'])}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Ga8gpNslpfDF"
   },
   "source": [
    "# 5. Update Config For Transfer Learning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "id": "Z9hRrO_ppfDF"
   },
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "from object_detection.utils import config_util\n",
    "from object_detection.protos import pipeline_pb2\n",
    "from google.protobuf import text_format"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "id": "c2A0mn4ipfDF"
   },
   "outputs": [],
   "source": [
    "config = config_util.get_configs_from_pipeline_file(files['PIPELINE_CONFIG'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "uQA13-afpfDF",
    "outputId": "907496a4-a39d-4b13-8c2c-e5978ecb1f10"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'model': ssd {\n",
       "   num_classes: 90\n",
       "   image_resizer {\n",
       "     fixed_shape_resizer {\n",
       "       height: 320\n",
       "       width: 320\n",
       "     }\n",
       "   }\n",
       "   feature_extractor {\n",
       "     type: \"ssd_mobilenet_v2_fpn_keras\"\n",
       "     depth_multiplier: 1.0\n",
       "     min_depth: 16\n",
       "     conv_hyperparams {\n",
       "       regularizer {\n",
       "         l2_regularizer {\n",
       "           weight: 4e-05\n",
       "         }\n",
       "       }\n",
       "       initializer {\n",
       "         random_normal_initializer {\n",
       "           mean: 0.0\n",
       "           stddev: 0.01\n",
       "         }\n",
       "       }\n",
       "       activation: RELU_6\n",
       "       batch_norm {\n",
       "         decay: 0.997\n",
       "         scale: true\n",
       "         epsilon: 0.001\n",
       "       }\n",
       "     }\n",
       "     use_depthwise: true\n",
       "     override_base_feature_extractor_hyperparams: true\n",
       "     fpn {\n",
       "       min_level: 3\n",
       "       max_level: 7\n",
       "       additional_layer_depth: 128\n",
       "     }\n",
       "   }\n",
       "   box_coder {\n",
       "     faster_rcnn_box_coder {\n",
       "       y_scale: 10.0\n",
       "       x_scale: 10.0\n",
       "       height_scale: 5.0\n",
       "       width_scale: 5.0\n",
       "     }\n",
       "   }\n",
       "   matcher {\n",
       "     argmax_matcher {\n",
       "       matched_threshold: 0.5\n",
       "       unmatched_threshold: 0.5\n",
       "       ignore_thresholds: false\n",
       "       negatives_lower_than_unmatched: true\n",
       "       force_match_for_each_row: true\n",
       "       use_matmul_gather: true\n",
       "     }\n",
       "   }\n",
       "   similarity_calculator {\n",
       "     iou_similarity {\n",
       "     }\n",
       "   }\n",
       "   box_predictor {\n",
       "     weight_shared_convolutional_box_predictor {\n",
       "       conv_hyperparams {\n",
       "         regularizer {\n",
       "           l2_regularizer {\n",
       "             weight: 4e-05\n",
       "           }\n",
       "         }\n",
       "         initializer {\n",
       "           random_normal_initializer {\n",
       "             mean: 0.0\n",
       "             stddev: 0.01\n",
       "           }\n",
       "         }\n",
       "         activation: RELU_6\n",
       "         batch_norm {\n",
       "           decay: 0.997\n",
       "           scale: true\n",
       "           epsilon: 0.001\n",
       "         }\n",
       "       }\n",
       "       depth: 128\n",
       "       num_layers_before_predictor: 4\n",
       "       kernel_size: 3\n",
       "       class_prediction_bias_init: -4.6\n",
       "       share_prediction_tower: true\n",
       "       use_depthwise: true\n",
       "     }\n",
       "   }\n",
       "   anchor_generator {\n",
       "     multiscale_anchor_generator {\n",
       "       min_level: 3\n",
       "       max_level: 7\n",
       "       anchor_scale: 4.0\n",
       "       aspect_ratios: 1.0\n",
       "       aspect_ratios: 2.0\n",
       "       aspect_ratios: 0.5\n",
       "       scales_per_octave: 2\n",
       "     }\n",
       "   }\n",
       "   post_processing {\n",
       "     batch_non_max_suppression {\n",
       "       score_threshold: 1e-08\n",
       "       iou_threshold: 0.6\n",
       "       max_detections_per_class: 100\n",
       "       max_total_detections: 100\n",
       "       use_static_shapes: false\n",
       "     }\n",
       "     score_converter: SIGMOID\n",
       "   }\n",
       "   normalize_loss_by_num_matches: true\n",
       "   loss {\n",
       "     localization_loss {\n",
       "       weighted_smooth_l1 {\n",
       "       }\n",
       "     }\n",
       "     classification_loss {\n",
       "       weighted_sigmoid_focal {\n",
       "         gamma: 2.0\n",
       "         alpha: 0.25\n",
       "       }\n",
       "     }\n",
       "     classification_weight: 1.0\n",
       "     localization_weight: 1.0\n",
       "   }\n",
       "   encode_background_as_zeros: true\n",
       "   normalize_loc_loss_by_codesize: true\n",
       "   inplace_batchnorm_update: true\n",
       "   freeze_batchnorm: false\n",
       " },\n",
       " 'train_config': batch_size: 128\n",
       " data_augmentation_options {\n",
       "   random_horizontal_flip {\n",
       "   }\n",
       " }\n",
       " data_augmentation_options {\n",
       "   random_crop_image {\n",
       "     min_object_covered: 0.0\n",
       "     min_aspect_ratio: 0.75\n",
       "     max_aspect_ratio: 3.0\n",
       "     min_area: 0.75\n",
       "     max_area: 1.0\n",
       "     overlap_thresh: 0.0\n",
       "   }\n",
       " }\n",
       " sync_replicas: true\n",
       " optimizer {\n",
       "   momentum_optimizer {\n",
       "     learning_rate {\n",
       "       cosine_decay_learning_rate {\n",
       "         learning_rate_base: 0.08\n",
       "         total_steps: 50000\n",
       "         warmup_learning_rate: 0.026666\n",
       "         warmup_steps: 1000\n",
       "       }\n",
       "     }\n",
       "     momentum_optimizer_value: 0.9\n",
       "   }\n",
       "   use_moving_average: false\n",
       " }\n",
       " fine_tune_checkpoint: \"PATH_TO_BE_CONFIGURED\"\n",
       " num_steps: 50000\n",
       " startup_delay_steps: 0.0\n",
       " replicas_to_aggregate: 8\n",
       " max_number_of_boxes: 100\n",
       " unpad_groundtruth_tensors: false\n",
       " fine_tune_checkpoint_type: \"classification\"\n",
       " fine_tune_checkpoint_version: V2,\n",
       " 'train_input_config': label_map_path: \"PATH_TO_BE_CONFIGURED\"\n",
       " tf_record_input_reader {\n",
       "   input_path: \"PATH_TO_BE_CONFIGURED\"\n",
       " },\n",
       " 'eval_config': metrics_set: \"coco_detection_metrics\"\n",
       " use_moving_averages: false,\n",
       " 'eval_input_configs': [label_map_path: \"PATH_TO_BE_CONFIGURED\"\n",
       " shuffle: false\n",
       " num_epochs: 1\n",
       " tf_record_input_reader {\n",
       "   input_path: \"PATH_TO_BE_CONFIGURED\"\n",
       " }\n",
       " ],\n",
       " 'eval_input_config': label_map_path: \"PATH_TO_BE_CONFIGURED\"\n",
       " shuffle: false\n",
       " num_epochs: 1\n",
       " tf_record_input_reader {\n",
       "   input_path: \"PATH_TO_BE_CONFIGURED\"\n",
       " }}"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "config"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "id": "9vK5lotDpfDF"
   },
   "outputs": [],
   "source": [
    "pipeline_config = pipeline_pb2.TrainEvalPipelineConfig()\n",
    "with tf.io.gfile.GFile(files['PIPELINE_CONFIG'], \"r\") as f:                                                                                                                                                                                                                     \n",
    "    proto_str = f.read()                                                                                                                                                                                                                                          \n",
    "    text_format.Merge(proto_str, pipeline_config)  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "id": "rP43Ph0JpfDG"
   },
   "outputs": [],
   "source": [
    "pipeline_config.model.ssd.num_classes = len(labels)\n",
    "pipeline_config.train_config.batch_size = 4\n",
    "pipeline_config.train_config.fine_tune_checkpoint = os.path.join(paths['PRETRAINED_MODEL_PATH'], PRETRAINED_MODEL_NAME, 'checkpoint', 'ckpt-0')\n",
    "pipeline_config.train_config.fine_tune_checkpoint_type = \"detection\"\n",
    "pipeline_config.train_input_reader.label_map_path= files['LABELMAP']\n",
    "pipeline_config.train_input_reader.tf_record_input_reader.input_path[:] = [os.path.join(paths['ANNOTATION_PATH'], 'train.record')]\n",
    "pipeline_config.eval_input_reader[0].label_map_path = files['LABELMAP']\n",
    "pipeline_config.eval_input_reader[0].tf_record_input_reader.input_path[:] = [os.path.join(paths['ANNOTATION_PATH'], 'test.record')]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "id": "oJvfgwWqpfDG"
   },
   "outputs": [],
   "source": [
    "config_text = text_format.MessageToString(pipeline_config)                                                                                                                                                                                                        \n",
    "with tf.io.gfile.GFile(files['PIPELINE_CONFIG'], \"wb\") as f:                                                                                                                                                                                                                     \n",
    "    f.write(config_text)   "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Zr3ON7xMpfDG"
   },
   "source": [
    "# 6. Train the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "id": "B-Y2UQmQpfDG"
   },
   "outputs": [],
   "source": [
    "TRAINING_SCRIPT = os.path.join(paths['APIMODEL_PATH'], 'research', 'object_detection', 'model_main_tf2.py')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "id": "jMP2XDfQpfDH"
   },
   "outputs": [],
   "source": [
    "command = \"python {} --model_dir={} --pipeline_config_path={} --num_train_steps=10000\".format(TRAINING_SCRIPT, paths['CHECKPOINT_PATH'],files['PIPELINE_CONFIG'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "A4OXXi-ApfDH",
    "outputId": "117a0e83-012b-466e-b7a6-ccaa349ac5ab"
   },
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'command' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "Cell \u001b[1;32mIn[60], line 1\u001b[0m\n\u001b[1;32m----> 1\u001b[0m \u001b[38;5;28mprint\u001b[39m(\u001b[43mcommand\u001b[49m)\n",
      "\u001b[1;31mNameError\u001b[0m: name 'command' is not defined"
     ]
    }
   ],
   "source": [
    "print(command)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "i3ZsJR-qpfDH",
    "outputId": "cabec5e1-45e6-4f2f-d9cf-297d9c1d0225"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "'{command}' is not recognized as an internal or external command,\n",
      "operable program or batch file.\n"
     ]
    }
   ],
   "source": [
    "!{command}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "4_YRZu7npfDH"
   },
   "source": [
    "# 7. Evaluate the Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "id": "80L7-fdPpfDH"
   },
   "outputs": [],
   "source": [
    "command = \"python {} --model_dir={} --pipeline_config_path={} --checkpoint_dir={}\".format(TRAINING_SCRIPT, paths['CHECKPOINT_PATH'],files['PIPELINE_CONFIG'], paths['CHECKPOINT_PATH'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "lYsgEPx9pfDH",
    "outputId": "8632d48b-91d2-45d9-bcb8-c1b172bf6eed"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "python Tensorflow\\models\\research\\object_detection\\model_main_tf2.py --model_dir=Tensorflow\\workspace\\models\\my_ssd_mobnet --pipeline_config_path=Tensorflow\\workspace\\models\\my_ssd_mobnet\\pipeline.config --checkpoint_dir=Tensorflow\\workspace\\models\\my_ssd_mobnet\n"
     ]
    }
   ],
   "source": [
    "print(command)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "lqTV2jGBpfDH"
   },
   "outputs": [],
   "source": [
    "!{command}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "orvRk02UpfDI"
   },
   "source": [
    "# 8. Load Train Model From Checkpoint"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "id": "8TYk4_oIpfDI"
   },
   "outputs": [],
   "source": [
    "import os\n",
    "import tensorflow as tf\n",
    "from object_detection.utils import label_map_util\n",
    "from object_detection.utils import visualization_utils as viz_utils\n",
    "from object_detection.builders import model_builder\n",
    "from object_detection.utils import config_util"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "# prevent GPU complete consumption\n",
    "gpus = tf.config.list_physical_devices('GPU')\n",
    "if gpus:\n",
    "    try:\n",
    "        tf.config.experimental.set_virtual_device_configuration(\n",
    "            gpus[0], [tf.config.experimental.VirtualDeviceConfiguration(memory_limit=5120)])\n",
    "    except RunTimeError as e:\n",
    "            print(e)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "id": "tDnQg-cYpfDI"
   },
   "outputs": [],
   "source": [
    "# Load pipeline config and build a detection model\n",
    "configs = config_util.get_configs_from_pipeline_file(files['PIPELINE_CONFIG'])\n",
    "detection_model = model_builder.build(model_config=configs['model'], is_training=False)\n",
    "\n",
    "# Restore checkpoint\n",
    "ckpt = tf.compat.v2.train.Checkpoint(model=detection_model)\n",
    "ckpt.restore(os.path.join(paths['CHECKPOINT_PATH'], 'ckpt-11')).expect_partial()\n",
    "\n",
    "@tf.function\n",
    "def detect_fn(image):\n",
    "    image, shapes = detection_model.preprocess(image)\n",
    "    prediction_dict = detection_model.predict(image, shapes)\n",
    "    detections = detection_model.postprocess(prediction_dict, shapes)\n",
    "    return detections"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "0EmsmbBZpfDI"
   },
   "source": [
    "# 9. Detect from an Image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "id": "Y_MKiuZ4pfDI"
   },
   "outputs": [],
   "source": [
    "import cv2 \n",
    "import numpy as np\n",
    "from matplotlib import pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: opencv-python in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (4.9.0.80)\n",
      "Requirement already satisfied: numpy>=1.21.2 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from opencv-python) (1.24.0)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\absl_py-2.0.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\apache_beam-2.53.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\avro_python3-1.10.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\contextlib2-21.6.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\cython-3.0.8-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\dm_tree-0.1.8-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\h5py-3.10.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lvis-0.5.3-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lxml-5.1.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\namex-0.0.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\object_detection-0.1-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pandas-2.2.0rc0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pillow-10.2.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\portalocker-2.8.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pycocotools-2.0.7-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pyparsing-2.4.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\regex-2023.12.25-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\rich-13.7.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\sacrebleu-2.2.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\scipy-1.12.0rc2-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tabulate-0.9.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io_gcs_filesystem-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_models_official-2.15.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_slim-1.1.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting opencv-contrib-python\n",
      "  Using cached opencv_contrib_python-4.9.0.80-cp37-abi3-win_amd64.whl.metadata (20 kB)\n",
      "Requirement already satisfied: numpy>=1.21.2 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from opencv-contrib-python) (1.24.0)\n",
      "Using cached opencv_contrib_python-4.9.0.80-cp37-abi3-win_amd64.whl (45.3 MB)\n",
      "Installing collected packages: opencv-contrib-python\n",
      "Successfully installed opencv-contrib-python-4.9.0.80\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\absl_py-2.0.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\apache_beam-2.53.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\avro_python3-1.10.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\contextlib2-21.6.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\cython-3.0.8-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\dm_tree-0.1.8-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\h5py-3.10.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lvis-0.5.3-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lxml-5.1.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\namex-0.0.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\object_detection-0.1-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pandas-2.2.0rc0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pillow-10.2.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\portalocker-2.8.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pycocotools-2.0.7-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pyparsing-2.4.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\regex-2023.12.25-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\rich-13.7.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\sacrebleu-2.2.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\scipy-1.12.0rc2-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tabulate-0.9.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io_gcs_filesystem-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_models_official-2.15.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_slim-1.1.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n"
     ]
    }
   ],
   "source": [
    "!pip install opencv-python\n",
    "!pip install opencv-contrib-python"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "id": "cBDbIhNapfDI"
   },
   "outputs": [],
   "source": [
    "category_index = label_map_util.create_category_index_from_labelmap(files['LABELMAP'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "id": "Lx3crOhOzITB"
   },
   "outputs": [],
   "source": [
    "IMAGE_PATH = os.path.join(paths['IMAGE_PATH'], 'test', 'Cars425.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING: Ignoring invalid distribution ~atplotlib (C:\\Users\\Hasitha\\Desktop\\ANPR\\anprsys\\Lib\\site-packages)\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\absl_py-2.0.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\apache_beam-2.53.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\avro_python3-1.10.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\contextlib2-21.6.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\cython-3.0.8-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\dm_tree-0.1.8-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\h5py-3.10.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lvis-0.5.3-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lxml-5.1.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\namex-0.0.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\object_detection-0.1-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pandas-2.2.0rc0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pillow-10.2.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\portalocker-2.8.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pycocotools-2.0.7-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pyparsing-2.4.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\regex-2023.12.25-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\rich-13.7.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\sacrebleu-2.2.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\scipy-1.12.0rc2-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tabulate-0.9.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io_gcs_filesystem-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_models_official-2.15.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_slim-1.1.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "WARNING: Skipping matplotlib as it is not installed.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting matplotlib\n",
      "  Using cached matplotlib-3.8.2-cp311-cp311-win_amd64.whl.metadata (5.9 kB)\n",
      "Collecting contourpy>=1.0.1 (from matplotlib)\n",
      "  Using cached contourpy-1.2.0-cp311-cp311-win_amd64.whl.metadata (5.8 kB)\n",
      "Requirement already satisfied: cycler>=0.10 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from matplotlib) (0.12.1)\n",
      "Collecting fonttools>=4.22.0 (from matplotlib)\n",
      "  Using cached fonttools-4.47.2-cp311-cp311-win_amd64.whl.metadata (160 kB)\n",
      "Requirement already satisfied: kiwisolver>=1.3.1 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from matplotlib) (1.4.5)\n",
      "Requirement already satisfied: numpy<2,>=1.21 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from matplotlib) (1.24.0)\n",
      "Requirement already satisfied: packaging>=20.0 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from matplotlib) (23.2)\n",
      "Requirement already satisfied: pillow>=8 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pillow-10.2.0-py3.11-win-amd64.egg (from matplotlib) (10.2.0)\n",
      "Requirement already satisfied: pyparsing>=2.3.1 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pyparsing-2.4.7-py3.11.egg (from matplotlib) (2.4.7)\n",
      "Requirement already satisfied: python-dateutil>=2.7 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from matplotlib) (2.8.2)\n",
      "Requirement already satisfied: six>=1.5 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from python-dateutil>=2.7->matplotlib) (1.16.0)\n",
      "Using cached matplotlib-3.8.2-cp311-cp311-win_amd64.whl (7.6 MB)\n",
      "Using cached contourpy-1.2.0-cp311-cp311-win_amd64.whl (187 kB)\n",
      "Using cached fonttools-4.47.2-cp311-cp311-win_amd64.whl (2.2 MB)\n",
      "Installing collected packages: fonttools, contourpy, matplotlib\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING: Ignoring invalid distribution ~atplotlib (C:\\Users\\Hasitha\\Desktop\\ANPR\\anprsys\\Lib\\site-packages)\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\absl_py-2.0.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\apache_beam-2.53.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\avro_python3-1.10.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\contextlib2-21.6.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\cython-3.0.8-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\dm_tree-0.1.8-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\h5py-3.10.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lvis-0.5.3-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lxml-5.1.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\namex-0.0.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\object_detection-0.1-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pandas-2.2.0rc0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pillow-10.2.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\portalocker-2.8.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pycocotools-2.0.7-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pyparsing-2.4.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\regex-2023.12.25-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\rich-13.7.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\sacrebleu-2.2.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\scipy-1.12.0rc2-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tabulate-0.9.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io_gcs_filesystem-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_models_official-2.15.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_slim-1.1.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "WARNING: Ignoring invalid distribution ~atplotlib (C:\\Users\\Hasitha\\Desktop\\ANPR\\anprsys\\Lib\\site-packages)\n",
      "ERROR: Could not install packages due to an OSError: [WinError 5] Access is denied: 'C:\\\\Users\\\\Hasitha\\\\Desktop\\\\ANPR\\\\anprsys\\\\Lib\\\\site-packages\\\\matplotlib\\\\ft2font.cp311-win_amd64.pyd'\n",
      "Check the permissions.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "!pip uninstall matplotlib\n",
    "!pip install matplotlib"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 269
    },
    "id": "Tpzn1SMry1yK",
    "outputId": "c392a2c5-10fe-4fc4-9998-a1d4c7db2bd3"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "img = cv2.imread(IMAGE_PATH)\n",
    "image_np = np.array(img)\n",
    "\n",
    "input_tensor = tf.convert_to_tensor(np.expand_dims(image_np, 0), dtype=tf.float32)\n",
    "detections = detect_fn(input_tensor)\n",
    "\n",
    "num_detections = int(detections.pop('num_detections'))\n",
    "detections = {key: value[0, :num_detections].numpy()\n",
    "              for key, value in detections.items()}\n",
    "detections['num_detections'] = num_detections\n",
    "\n",
    "# detection_classes should be ints.\n",
    "detections['detection_classes'] = detections['detection_classes'].astype(np.int64)\n",
    "\n",
    "label_id_offset = 1\n",
    "image_np_with_detections = image_np.copy()\n",
    "\n",
    "viz_utils.visualize_boxes_and_labels_on_image_array(\n",
    "            image_np_with_detections,\n",
    "            detections['detection_boxes'],\n",
    "            detections['detection_classes']+label_id_offset,\n",
    "            detections['detection_scores'],\n",
    "            category_index,\n",
    "            use_normalized_coordinates=True,\n",
    "            max_boxes_to_draw=5,\n",
    "            min_score_thresh=.8,\n",
    "            agnostic_mode=False)\n",
    "\n",
    "plt.imshow(cv2.cvtColor(image_np_with_detections, cv2.COLOR_BGR2RGB))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['detection_boxes', 'detection_scores', 'detection_classes', 'raw_detection_boxes', 'raw_detection_scores', 'detection_multiclass_scores', 'detection_anchor_indices', 'num_detections'])"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "detections.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Implement EasyOCR"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting easyocr\n",
      "  Using cached easyocr-1.7.1-py3-none-any.whl.metadata (11 kB)\n",
      "Requirement already satisfied: torch in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from easyocr) (2.1.0+cu118)\n",
      "Requirement already satisfied: torchvision>=0.5 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from easyocr) (0.16.0+cu118)\n",
      "Collecting opencv-python-headless (from easyocr)\n",
      "  Using cached opencv_python_headless-4.9.0.80-cp37-abi3-win_amd64.whl.metadata (20 kB)\n",
      "Requirement already satisfied: scipy in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\scipy-1.12.0rc2-py3.11-win-amd64.egg (from easyocr) (1.12.0rc2)\n",
      "Requirement already satisfied: numpy in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from easyocr) (1.24.0)\n",
      "Requirement already satisfied: Pillow in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pillow-10.2.0-py3.11-win-amd64.egg (from easyocr) (10.2.0)\n",
      "Collecting scikit-image (from easyocr)\n",
      "  Using cached scikit_image-0.22.0-cp311-cp311-win_amd64.whl.metadata (13 kB)\n",
      "Requirement already satisfied: python-bidi in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from easyocr) (0.4.2)\n",
      "Requirement already satisfied: PyYAML in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from easyocr) (6.0.1)\n",
      "Requirement already satisfied: Shapely in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from easyocr) (2.0.2)\n",
      "Requirement already satisfied: pyclipper in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from easyocr) (1.3.0.post5)\n",
      "Requirement already satisfied: ninja in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from easyocr) (1.11.1.1)\n",
      "Requirement already satisfied: requests in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from torchvision>=0.5->easyocr) (2.31.0)\n",
      "Requirement already satisfied: filelock in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from torch->easyocr) (3.9.0)\n",
      "Requirement already satisfied: typing-extensions in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from torch->easyocr) (4.5.0)\n",
      "Requirement already satisfied: sympy in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from torch->easyocr) (1.12)\n",
      "Requirement already satisfied: networkx in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from torch->easyocr) (3.0)\n",
      "Requirement already satisfied: jinja2 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from torch->easyocr) (3.1.3)\n",
      "Requirement already satisfied: fsspec in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from torch->easyocr) (2023.4.0)\n",
      "Requirement already satisfied: six in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from python-bidi->easyocr) (1.16.0)\n",
      "Collecting imageio>=2.27 (from scikit-image->easyocr)\n",
      "  Using cached imageio-2.33.1-py3-none-any.whl.metadata (4.9 kB)\n",
      "Requirement already satisfied: tifffile>=2022.8.12 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from scikit-image->easyocr) (2023.12.9)\n",
      "Requirement already satisfied: packaging>=21 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from scikit-image->easyocr) (23.2)\n",
      "Collecting lazy_loader>=0.3 (from scikit-image->easyocr)\n",
      "  Using cached lazy_loader-0.3-py3-none-any.whl.metadata (4.3 kB)\n",
      "Requirement already satisfied: MarkupSafe>=2.0 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from jinja2->torch->easyocr) (2.1.3)\n",
      "Requirement already satisfied: charset-normalizer<4,>=2 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from requests->torchvision>=0.5->easyocr) (3.3.2)\n",
      "Requirement already satisfied: idna<4,>=2.5 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from requests->torchvision>=0.5->easyocr) (3.6)\n",
      "Requirement already satisfied: urllib3<3,>=1.21.1 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from requests->torchvision>=0.5->easyocr) (2.1.0)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from requests->torchvision>=0.5->easyocr) (2023.11.17)\n",
      "Requirement already satisfied: mpmath>=0.19 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from sympy->torch->easyocr) (1.3.0)\n",
      "Using cached easyocr-1.7.1-py3-none-any.whl (2.9 MB)\n",
      "Using cached opencv_python_headless-4.9.0.80-cp37-abi3-win_amd64.whl (38.5 MB)\n",
      "Using cached scikit_image-0.22.0-cp311-cp311-win_amd64.whl (24.5 MB)\n",
      "Using cached imageio-2.33.1-py3-none-any.whl (313 kB)\n",
      "Using cached lazy_loader-0.3-py3-none-any.whl (9.1 kB)\n",
      "Installing collected packages: opencv-python-headless, lazy_loader, imageio, scikit-image, easyocr\n",
      "Successfully installed easyocr-1.7.1 imageio-2.33.1 lazy_loader-0.3 opencv-python-headless-4.9.0.80 scikit-image-0.22.0\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\absl_py-2.0.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\apache_beam-2.53.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\avro_python3-1.10.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\contextlib2-21.6.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\cython-3.0.8-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\dm_tree-0.1.8-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\h5py-3.10.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lvis-0.5.3-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lxml-5.1.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\namex-0.0.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\object_detection-0.1-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pandas-2.2.0rc0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pillow-10.2.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\portalocker-2.8.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pycocotools-2.0.7-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pyparsing-2.4.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\regex-2023.12.25-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\rich-13.7.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\sacrebleu-2.2.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\scipy-1.12.0rc2-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tabulate-0.9.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io_gcs_filesystem-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_models_official-2.15.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_slim-1.1.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n",
      "tf-models-official 2.15.0 requires google-api-python-client>=1.6.7, which is not installed.\n",
      "tf-models-official 2.15.0 requires immutabledict, which is not installed.\n",
      "tf-models-official 2.15.0 requires kaggle>=1.3.9, which is not installed.\n",
      "tf-models-official 2.15.0 requires oauth2client, which is not installed.\n",
      "tf-models-official 2.15.0 requires py-cpuinfo>=3.3.0, which is not installed.\n",
      "tf-models-official 2.15.0 requires sentencepiece, which is not installed.\n",
      "tf-models-official 2.15.0 requires seqeval, which is not installed.\n",
      "tf-models-official 2.15.0 requires tensorflow-datasets, which is not installed.\n",
      "tf-models-official 2.15.0 requires tensorflow-hub>=0.6.0, which is not installed.\n",
      "tf-models-official 2.15.0 requires tensorflow-model-optimization>=0.4.1, which is not installed.\n",
      "tf-models-official 2.15.0 requires tensorflow-text~=2.15.0, which is not installed.\n",
      "tf-models-official 2.15.0 requires tensorflow~=2.15.0, but you have tensorflow 2.13.0 which is incompatible.\n"
     ]
    }
   ],
   "source": [
    "!pip3 install easyocr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Looking in indexes: https://download.pytorch.org/whl/cu118\n",
      "Requirement already satisfied: torch==2.1.0 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (2.1.0+cu118)\n",
      "Requirement already satisfied: torchvision==0.16.0 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (0.16.0+cu118)\n",
      "Requirement already satisfied: torchaudio==2.1.0 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (2.1.0+cu118)\n",
      "Requirement already satisfied: filelock in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from torch==2.1.0) (3.9.0)\n",
      "Requirement already satisfied: typing-extensions in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from torch==2.1.0) (4.5.0)\n",
      "Requirement already satisfied: sympy in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from torch==2.1.0) (1.12)\n",
      "Requirement already satisfied: networkx in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from torch==2.1.0) (3.0)\n",
      "Requirement already satisfied: jinja2 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from torch==2.1.0) (3.1.3)\n",
      "Requirement already satisfied: fsspec in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from torch==2.1.0) (2023.4.0)\n",
      "Requirement already satisfied: numpy in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from torchvision==0.16.0) (1.24.0)\n",
      "Requirement already satisfied: requests in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from torchvision==0.16.0) (2.31.0)\n",
      "Requirement already satisfied: pillow!=8.3.*,>=5.3.0 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pillow-10.2.0-py3.11-win-amd64.egg (from torchvision==0.16.0) (10.2.0)\n",
      "Requirement already satisfied: MarkupSafe>=2.0 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from jinja2->torch==2.1.0) (2.1.3)\n",
      "Requirement already satisfied: charset-normalizer<4,>=2 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from requests->torchvision==0.16.0) (3.3.2)\n",
      "Requirement already satisfied: idna<4,>=2.5 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from requests->torchvision==0.16.0) (3.6)\n",
      "Requirement already satisfied: urllib3<3,>=1.21.1 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from requests->torchvision==0.16.0) (2.1.0)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from requests->torchvision==0.16.0) (2023.11.17)\n",
      "Requirement already satisfied: mpmath>=0.19 in c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages (from sympy->torch==2.1.0) (1.3.0)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\absl_py-2.0.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\apache_beam-2.53.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\avro_python3-1.10.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\contextlib2-21.6.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\cython-3.0.8-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\dm_tree-0.1.8-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\h5py-3.10.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lvis-0.5.3-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\lxml-5.1.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\namex-0.0.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\object_detection-0.1-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pandas-2.2.0rc0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pillow-10.2.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\portalocker-2.8.2-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pycocotools-2.0.7-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\pyparsing-2.4.7-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\regex-2023.12.25-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\rich-13.7.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\sacrebleu-2.2.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\scipy-1.12.0rc2-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tabulate-0.9.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tensorflow_io_gcs_filesystem-0.31.0-py3.11-win-amd64.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_models_official-2.15.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n",
      "DEPRECATION: Loading egg at c:\\users\\hasitha\\desktop\\anpr\\anprsys\\lib\\site-packages\\tf_slim-1.1.0-py3.11.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\n"
     ]
    }
   ],
   "source": [
    "!pip3 install torch==2.1.0 torchvision==0.16.0 torchaudio==2.1.0 --index-url https://download.pytorch.org/whl/cu118"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "import easyocr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "detection_threshold = 0.4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "image = image_np_with_detections\n",
    "scores = list(filter(lambda x: x> detection_threshold, detections['detection_scores']))\n",
    "boxes = detections['detection_boxes'][:len(scores)]\n",
    "classes = detections['detection_classes'][:len(scores)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ True, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False])"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "detections['detection_scores'] > detection_threshold"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.44893554, 0.41207653, 0.5883073 , 0.6108685 ]], dtype=float32)"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "boxes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "width = image.shape[1]\n",
    "height = image.shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "232"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "height"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[([[8, 4], [90, 4], [90, 28], [8, 28]], '6526 JHD', 0.9231559254521261)]\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Apply ROI filtering with OCR\n",
    "\n",
    "for idx, box in enumerate(boxes):\n",
    "    roi = box*[height, width, height, width]\n",
    "    region = image[int(roi[0]):int(roi[2]),int(roi[1]):int(roi[3])]\n",
    "    reader = easyocr.Reader(['en'])\n",
    "    ocr_result = reader.readtext(region)\n",
    "    print(ocr_result)\n",
    "    plt.imshow(cv2.cvtColor(region, cv2.COLOR_BGR2RGB))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "region_threshold = 0.6    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "def filter_text(region, ocr_result, region_threshold):\n",
    "    rectangle_size = region.shape[0]*region.shape[1]\n",
    "    plate = []\n",
    "    for result in ocr_result:\n",
    "        length = np.sum(np.subtract(result[0][1], result[0][0]))\n",
    "        height = np.sum(np.subtract(result[0][2], result[0][1]))\n",
    "        if length*height / rectangle_size > region_threshold:\n",
    "            plate.append(result[1])\n",
    "        print(length, width)\n",
    "\n",
    "    return plate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "82 500\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "['6526 JHD']"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filter_text(region, ocr_result, region_threshold)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "def ocr_it(image, detections,detection_threshold, region_threshold):\n",
    "    # scores boxes and classes above threshold\n",
    "    scores = list(filter(lambda x: x> detection_threshold, detections['detection_scores']))\n",
    "    boxes = detections['detection_boxes'][:len(scores)]\n",
    "    classes = detections['detection_classes'][:len(scores)]\n",
    "\n",
    "    # Full Image Dimention\n",
    "    width = image.shape[1]\n",
    "    height = image.shape[0]\n",
    "\n",
    "    # Apply ROI filtering with OCR\n",
    "\n",
    "    for idx, box in enumerate(boxes):\n",
    "        roi = box*[height, width, height, width]\n",
    "        region = image[int(roi[0]):int(roi[2]),int(roi[1]):int(roi[3])]\n",
    "        reader = easyocr.Reader(['en'])\n",
    "        ocr_result = reader.readtext(region)\n",
    "\n",
    "        text = filter_text(region, ocr_result, region_threshold)\n",
    "        \n",
    "        plt.imshow(cv2.cvtColor(region, cv2.COLOR_BGR2RGB))\n",
    "        print(text)\n",
    "        return text, region"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "82 500\n",
      "['6526 JHD']\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "text, region = ocr_it(image_np_with_detections, detections, detection_threshold, region_threshold)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "IsNAaYAo0WVL"
   },
   "source": [
    "# 10. Real Time Detections from your Webcam"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "Usage:   \n",
      "  pip install [options] <requirement specifier> [package-index-options] ...\n",
      "  pip install [options] -r <requirements file> [package-index-options] ...\n",
      "  pip install [options] [-e] <vcs project url> ...\n",
      "  pip install [options] [-e] <local project path> ...\n",
      "  pip install [options] <archive url/path> ...\n",
      "\n",
      "no such option: -y\n"
     ]
    }
   ],
   "source": [
    "!pip install opencv-python-headless -y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "id": "o_grs6OGpfDJ",
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[]\n",
      "122 640\n",
      "['BI2H']\n",
      "123 640\n",
      "['B123']\n",
      "124 640\n",
      "['BIDS']\n",
      "126 640\n",
      "['LBLHA1023']\n",
      "128 640\n",
      "['@BT1023']\n",
      "126 640\n",
      "['@B T1023|']\n",
      "130 640\n",
      "['@BT1023']\n",
      "128 640\n",
      "['LBTA1023']\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cap = cv2.VideoCapture(0)\n",
    "width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))\n",
    "height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))\n",
    "\n",
    "while cap.isOpened(): \n",
    "    ret, frame = cap.read()\n",
    "    image_np = np.array(frame)\n",
    "    \n",
    "    input_tensor = tf.convert_to_tensor(np.expand_dims(image_np, 0), dtype=tf.float32)\n",
    "    detections = detect_fn(input_tensor)\n",
    "    \n",
    "    num_detections = int(detections.pop('num_detections'))\n",
    "    detections = {key: value[0, :num_detections].numpy()\n",
    "                  for key, value in detections.items()}\n",
    "    detections['num_detections'] = num_detections\n",
    "\n",
    "    # detection_classes should be ints.\n",
    "    detections['detection_classes'] = detections['detection_classes'].astype(np.int64)\n",
    "\n",
    "    label_id_offset = 1\n",
    "    image_np_with_detections = image_np.copy()\n",
    "\n",
    "    viz_utils.visualize_boxes_and_labels_on_image_array(\n",
    "                image_np_with_detections,\n",
    "                detections['detection_boxes'],\n",
    "                detections['detection_classes']+label_id_offset,\n",
    "                detections['detection_scores'],\n",
    "                category_index,\n",
    "                use_normalized_coordinates=True,\n",
    "                max_boxes_to_draw=5,\n",
    "                min_score_thresh=.8,\n",
    "                agnostic_mode=False)\n",
    "    try:\n",
    "        text, region = ocr_it(image_np_with_detections, detections, detection_threshold, region_threshold)\n",
    "        save_results(text, region, 'realtime_images.csv', 'Detection_Images')\n",
    "    except: \n",
    "        pass\n",
    "\n",
    "    \n",
    "    cv2.imshow('object detection',  cv2.resize(image_np_with_detections, (800, 600)))\n",
    "    \n",
    "    if cv2.waitKey(10) & 0xFF == ord('q'):\n",
    "        cap.release()\n",
    "        cv2.destroyAllWindows()\n",
    "        break"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# save Result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "import csv\n",
    "import uuid"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'8d0cb7bf-b755-11ee-9eac-e00af68d7526.jpg'"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'{}.jpg'.format(uuid.uuid1())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "def save_results(text, region, csv_filename, folder_path):\n",
    "    img_name = '{}.jpg'.format(uuid.uuid1())\n",
    "\n",
    "    cv2.imwrite(os.path.join(folder_path, img_name), region)\n",
    "\n",
    "    with open(csv_filename, mode='a', newline='') as f:\n",
    "        csv_writer = csv.writer(f, delimiter=',', quotechar='\"', quoting=csv.QUOTE_MINIMAL)\n",
    "        csv_writer.writerow([img_name, text])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[[127, 255,   0],\n",
       "        [127, 255,   0],\n",
       "        [127, 255,   0],\n",
       "        ...,\n",
       "        [127, 255,   0],\n",
       "        [127, 255,   0],\n",
       "        [127, 255,   0]],\n",
       "\n",
       "       [[127, 255,   0],\n",
       "        [127, 255,   0],\n",
       "        [127, 255,   0],\n",
       "        ...,\n",
       "        [127, 255,   0],\n",
       "        [127, 255,   0],\n",
       "        [127, 255,   0]],\n",
       "\n",
       "       [[127, 255,   0],\n",
       "        [127, 255,   0],\n",
       "        [127, 255,   0],\n",
       "        ...,\n",
       "        [ 85,  37,  19],\n",
       "        [127, 255,   0],\n",
       "        [127, 255,   0]],\n",
       "\n",
       "       ...,\n",
       "\n",
       "       [[127, 255,   0],\n",
       "        [127, 255,   0],\n",
       "        [127, 255,   0],\n",
       "        ...,\n",
       "        [ 88, 130, 160],\n",
       "        [127, 255,   0],\n",
       "        [127, 255,   0]],\n",
       "\n",
       "       [[127, 255,   0],\n",
       "        [127, 255,   0],\n",
       "        [127, 255,   0],\n",
       "        ...,\n",
       "        [ 92, 127, 156],\n",
       "        [127, 255,   0],\n",
       "        [127, 255,   0]],\n",
       "\n",
       "       [[127, 255,   0],\n",
       "        [127, 255,   0],\n",
       "        [127, 255,   0],\n",
       "        ...,\n",
       "        [127, 255,   0],\n",
       "        [127, 255,   0],\n",
       "        [127, 255,   0]]], dtype=uint8)"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "region"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "save_results(text, region, 'detection_images.csv', 'Detection_Images')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "rzlM4jt0pfDJ"
   },
   "source": [
    "# 10. Freezing the Graph"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "n4olHB2npfDJ"
   },
   "outputs": [],
   "source": [
    "FREEZE_SCRIPT = os.path.join(paths['APIMODEL_PATH'], 'research', 'object_detection', 'exporter_main_v2.py ')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "0AjO93QDpfDJ"
   },
   "outputs": [],
   "source": [
    "command = \"python {} --input_type=image_tensor --pipeline_config_path={} --trained_checkpoint_dir={} --output_directory={}\".format(FREEZE_SCRIPT ,files['PIPELINE_CONFIG'], paths['CHECKPOINT_PATH'], paths['OUTPUT_PATH'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "F6Lsp3tCpfDJ",
    "outputId": "c3828529-bf06-4df5-d7f3-145890ec3edd"
   },
   "outputs": [],
   "source": [
    "print(command)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "1Sw1ULgHpfDJ",
    "outputId": "6fd441e1-9fc9-4889-d072-3395c21e40b6"
   },
   "outputs": [],
   "source": [
    "!{command}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "wTPmdqaXpfDK"
   },
   "source": [
    "# 11. Conversion to TFJS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "gZ6UzY_fpfDK",
    "outputId": "0c84722e-1c2b-4002-d857-80827ade828a",
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "!pip install tensorflowjs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "0oxbVynHpfDK"
   },
   "outputs": [],
   "source": [
    "command = \"tensorflowjs_converter --input_format=tf_saved_model --output_node_names='detection_boxes,detection_classes,detection_features,detection_multiclass_scores,detection_scores,num_detections,raw_detection_boxes,raw_detection_scores' --output_format=tfjs_graph_model --signature_name=serving_default {} {}\".format(os.path.join(paths['OUTPUT_PATH'], 'saved_model'), paths['TFJS_PATH'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "DB2AGNmJpfDK",
    "outputId": "fbc9f747-f511-47e8-df8f-5ea65cef0374"
   },
   "outputs": [],
   "source": [
    "print(command)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "K7rfT4-hpfDK",
    "outputId": "532707fd-6feb-4bc6-84a3-325b5d16303c"
   },
   "outputs": [],
   "source": [
    "!{command}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "o8_hm-itpfDK"
   },
   "outputs": [],
   "source": [
    "# Test Code: https://github.com/nicknochnack/RealTimeSignLanguageDetectionwithTFJS"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "VtUw73FHpfDK"
   },
   "source": [
    "# 12. Conversion to TFLite"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "XviMtewLpfDK"
   },
   "outputs": [],
   "source": [
    "TFLITE_SCRIPT = os.path.join(paths['APIMODEL_PATH'], 'research', 'object_detection', 'export_tflite_graph_tf2.py ')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "us86cjC4pfDL"
   },
   "outputs": [],
   "source": [
    "command = \"python {} --pipeline_config_path={} --trained_checkpoint_dir={} --output_directory={}\".format(TFLITE_SCRIPT ,files['PIPELINE_CONFIG'], paths['CHECKPOINT_PATH'], paths['TFLITE_PATH'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "n1r5YO3rpfDL",
    "outputId": "5fcdf7a4-eee2-4365-f1ca-1751968379ea"
   },
   "outputs": [],
   "source": [
    "print(command)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "I-xWpHN8pfDL",
    "outputId": "7f6bacd8-d077-43b5-c131-5b081fba24a4"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2021-04-03 11:55:14.884114: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1267]      0 \n",
      "2021-04-03 11:55:14.884117: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1280] 0:   N \n",
      "2021-04-03 11:55:14.884182: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1406] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 6611 MB memory) -> physical GPU (device: 0, name: GeForce RTX 2070 SUPER, pci bus id: 0000:2b:00.0, compute capability: 7.5)\n",
      "2021-04-03 11:55:14.884197: I tensorflow/compiler/jit/xla_gpu_device.cc:99] Not creating XLA devices, tf_xla_enable_xla_devices not set\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <object_detection.meta_architectures.ssd_meta_arch.SSDMetaArch object at 0x00000224AC2B1908>, because it is not built.\n",
      "W0403 11:55:15.360029 15552 save_impl.py:78] Skipping full serialization of Keras layer <object_detection.meta_architectures.ssd_meta_arch.SSDMetaArch object at 0x00000224AC2B1908>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <tensorflow.python.keras.layers.convolutional.SeparableConv2D object at 0x00000224D6D42B00>, because it is not built.\n",
      "W0403 11:55:15.925411 15552 save_impl.py:78] Skipping full serialization of Keras layer <tensorflow.python.keras.layers.convolutional.SeparableConv2D object at 0x00000224D6D42B00>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6CEA748>, because it is not built.\n",
      "W0403 11:55:15.925411 15552 save_impl.py:78] Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6CEA748>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D4763198>, because it is not built.\n",
      "W0403 11:55:15.925411 15552 save_impl.py:78] Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D4763198>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <tensorflow.python.keras.layers.convolutional.SeparableConv2D object at 0x00000224D4763D68>, because it is not built.\n",
      "W0403 11:55:15.925411 15552 save_impl.py:78] Skipping full serialization of Keras layer <tensorflow.python.keras.layers.convolutional.SeparableConv2D object at 0x00000224D4763D68>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6D01A90>, because it is not built.\n",
      "W0403 11:55:15.925411 15552 save_impl.py:78] Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6D01A90>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6D01358>, because it is not built.\n",
      "W0403 11:55:15.925411 15552 save_impl.py:78] Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6D01358>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <tensorflow.python.keras.layers.convolutional.SeparableConv2D object at 0x00000224D6D015C0>, because it is not built.\n",
      "W0403 11:55:15.925411 15552 save_impl.py:78] Skipping full serialization of Keras layer <tensorflow.python.keras.layers.convolutional.SeparableConv2D object at 0x00000224D6D015C0>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6CE36D8>, because it is not built.\n",
      "W0403 11:55:15.925411 15552 save_impl.py:78] Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6CE36D8>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6CE3CF8>, because it is not built.\n",
      "W0403 11:55:15.925411 15552 save_impl.py:78] Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6CE3CF8>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <tensorflow.python.keras.layers.convolutional.SeparableConv2D object at 0x00000224D6CE3F98>, because it is not built.\n",
      "W0403 11:55:15.925411 15552 save_impl.py:78] Skipping full serialization of Keras layer <tensorflow.python.keras.layers.convolutional.SeparableConv2D object at 0x00000224D6CE3F98>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6CB2470>, because it is not built.\n",
      "W0403 11:55:15.926413 15552 save_impl.py:78] Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6CB2470>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D477FDD8>, because it is not built.\n",
      "W0403 11:55:15.926413 15552 save_impl.py:78] Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D477FDD8>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6D57390>, because it is not built.\n",
      "W0403 11:55:15.926413 15552 save_impl.py:78] Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6D57390>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6D576D8>, because it is not built.\n",
      "W0403 11:55:15.926413 15552 save_impl.py:78] Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6D576D8>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6D6F390>, because it is not built.\n",
      "W0403 11:55:15.926413 15552 save_impl.py:78] Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6D6F390>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6D6F630>, because it is not built.\n",
      "W0403 11:55:15.926413 15552 save_impl.py:78] Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6D6F630>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6D6FCF8>, because it is not built.\n",
      "W0403 11:55:15.926413 15552 save_impl.py:78] Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6D6FCF8>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D4753320>, because it is not built.\n",
      "W0403 11:55:15.926413 15552 save_impl.py:78] Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D4753320>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D47534A8>, because it is not built.\n",
      "W0403 11:55:15.926413 15552 save_impl.py:78] Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D47534A8>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6D5F550>, because it is not built.\n",
      "W0403 11:55:15.926413 15552 save_impl.py:78] Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6D5F550>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D4774908>, because it is not built.\n",
      "W0403 11:55:15.926413 15552 save_impl.py:78] Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D4774908>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D46CBD68>, because it is not built.\n",
      "W0403 11:55:15.926413 15552 save_impl.py:78] Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D46CBD68>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D46CBDA0>, because it is not built.\n",
      "W0403 11:55:15.926413 15552 save_impl.py:78] Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D46CBDA0>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D46CB7B8>, because it is not built.\n",
      "W0403 11:55:15.926413 15552 save_impl.py:78] Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D46CB7B8>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D46CB8D0>, because it is not built.\n",
      "W0403 11:55:15.926413 15552 save_impl.py:78] Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D46CB8D0>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6D26828>, because it is not built.\n",
      "W0403 11:55:15.926413 15552 save_impl.py:78] Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6D26828>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6D26940>, because it is not built.\n",
      "W0403 11:55:15.926413 15552 save_impl.py:78] Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6D26940>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6D262E8>, because it is not built.\n",
      "W0403 11:55:15.926413 15552 save_impl.py:78] Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6D262E8>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6D12B70>, because it is not built.\n",
      "W0403 11:55:15.927411 15552 save_impl.py:78] Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6D12B70>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6D088D0>, because it is not built.\n",
      "W0403 11:55:15.927411 15552 save_impl.py:78] Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6D088D0>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6D08550>, because it is not built.\n",
      "W0403 11:55:15.927411 15552 save_impl.py:78] Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6D08550>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6D08F60>, because it is not built.\n",
      "W0403 11:55:15.927411 15552 save_impl.py:78] Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6D08F60>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6D08CF8>, because it is not built.\n",
      "W0403 11:55:15.927411 15552 save_impl.py:78] Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6D08CF8>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6D08710>, because it is not built.\n",
      "W0403 11:55:15.927411 15552 save_impl.py:78] Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6D08710>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6D086A0>, because it is not built.\n",
      "W0403 11:55:15.927411 15552 save_impl.py:78] Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6D086A0>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6D0C320>, because it is not built.\n",
      "W0403 11:55:15.927411 15552 save_impl.py:78] Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6D0C320>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6D72320>, because it is not built.\n",
      "W0403 11:55:15.927411 15552 save_impl.py:78] Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6D72320>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6D37748>, because it is not built.\n",
      "W0403 11:55:15.927411 15552 save_impl.py:78] Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6D37748>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6D37240>, because it is not built.\n",
      "W0403 11:55:15.927411 15552 save_impl.py:78] Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6D37240>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6D376A0>, because it is not built.\n",
      "W0403 11:55:15.927411 15552 save_impl.py:78] Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6D376A0>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6D37710>, because it is not built.\n",
      "W0403 11:55:15.927411 15552 save_impl.py:78] Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6D37710>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6D76198>, because it is not built.\n",
      "W0403 11:55:15.927411 15552 save_impl.py:78] Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6D76198>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6D76550>, because it is not built.\n",
      "W0403 11:55:15.927411 15552 save_impl.py:78] Skipping full serialization of Keras layer <object_detection.core.freezable_batch_norm.FreezableBatchNorm object at 0x00000224D6D76550>, because it is not built.\n",
      "WARNING:tensorflow:Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6D766A0>, because it is not built.\n",
      "W0403 11:55:15.927411 15552 save_impl.py:78] Skipping full serialization of Keras layer <tensorflow.python.keras.layers.core.Lambda object at 0x00000224D6D766A0>, because it is not built.\n",
      "2021-04-03 11:55:22.118760: W tensorflow/python/util/util.cc:348] Sets are not currently considered sequences, but this may change in the future, so consider avoiding using them.\n",
      "2021-04-03 11:55:23.021692: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1720] Found device 0 with properties: \n",
      "pciBusID: 0000:2b:00.0 name: GeForce RTX 2070 SUPER computeCapability: 7.5\n",
      "coreClock: 1.785GHz coreCount: 40 deviceMemorySize: 8.00GiB deviceMemoryBandwidth: 417.29GiB/s\n",
      "2021-04-03 11:55:23.021718: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cudart64_110.dll\n",
      "2021-04-03 11:55:23.021724: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cublas64_11.dll\n",
      "2021-04-03 11:55:23.021731: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cublasLt64_11.dll\n",
      "2021-04-03 11:55:23.021737: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cufft64_10.dll\n",
      "2021-04-03 11:55:23.021742: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library curand64_10.dll\n",
      "2021-04-03 11:55:23.021748: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cusolver64_10.dll\n",
      "2021-04-03 11:55:23.021754: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cusparse64_11.dll\n",
      "2021-04-03 11:55:23.021758: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cudnn64_8.dll\n",
      "2021-04-03 11:55:23.021783: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1862] Adding visible gpu devices: 0\n",
      "2021-04-03 11:55:23.021821: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1261] Device interconnect StreamExecutor with strength 1 edge matrix:\n",
      "2021-04-03 11:55:23.021825: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1267]      0 \n",
      "2021-04-03 11:55:23.021828: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1280] 0:   N \n",
      "2021-04-03 11:55:23.021886: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1406] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 6611 MB memory) -> physical GPU (device: 0, name: GeForce RTX 2070 SUPER, pci bus id: 0000:2b:00.0, compute capability: 7.5)\n",
      "2021-04-03 11:55:23.021901: I tensorflow/compiler/jit/xla_gpu_device.cc:99] Not creating XLA devices, tf_xla_enable_xla_devices not set\n",
      "INFO:tensorflow:Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C208>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C2E8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C550>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], False), {}).\n",
      "I0403 11:55:27.221577 15552 def_function.py:1170] Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C208>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C2E8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C550>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], False), {}).\n",
      "INFO:tensorflow:Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A860>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A5F8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A748>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], True), {}).\n",
      "I0403 11:55:27.221577 15552 def_function.py:1170] Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A860>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A5F8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A748>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], True), {}).\n",
      "INFO:tensorflow:Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A4F28>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A46D8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A4518>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], False), {}).\n",
      "I0403 11:55:27.222574 15552 def_function.py:1170] Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A4F28>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A46D8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A4518>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], False), {}).\n",
      "INFO:tensorflow:Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095EB8>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095C88>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095D68>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], True), {}).\n",
      "I0403 11:55:27.222574 15552 def_function.py:1170] Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095EB8>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095C88>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095D68>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], True), {}).\n",
      "INFO:tensorflow:Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A4F28>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A46D8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A4518>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], False), {}).\n",
      "I0403 11:55:29.965919 15552 def_function.py:1170] Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A4F28>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A46D8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A4518>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], False), {}).\n",
      "INFO:tensorflow:Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095EB8>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095C88>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095D68>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], True), {}).\n",
      "I0403 11:55:29.965919 15552 def_function.py:1170] Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095EB8>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095C88>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095D68>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], True), {}).\n",
      "INFO:tensorflow:Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C208>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C2E8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C550>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], False), {}).\n",
      "I0403 11:55:29.965919 15552 def_function.py:1170] Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C208>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C2E8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C550>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], False), {}).\n",
      "INFO:tensorflow:Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A860>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A5F8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A748>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], True), {}).\n",
      "I0403 11:55:29.965919 15552 def_function.py:1170] Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A860>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A5F8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A748>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], True), {}).\n",
      "INFO:tensorflow:Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C208>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C2E8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C550>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], False), {}).\n",
      "I0403 11:55:29.965919 15552 def_function.py:1170] Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C208>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C2E8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C550>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], False), {}).\n",
      "INFO:tensorflow:Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A860>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A5F8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A748>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], True), {}).\n",
      "I0403 11:55:29.966920 15552 def_function.py:1170] Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A860>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A5F8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A748>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], True), {}).\n",
      "W0403 11:55:32.109875 15552 save.py:241] Found untraced functions such as WeightSharedConvolutionalBoxPredictor_layer_call_and_return_conditional_losses, WeightSharedConvolutionalBoxPredictor_layer_call_fn, WeightSharedConvolutionalBoxHead_layer_call_and_return_conditional_losses, WeightSharedConvolutionalBoxHead_layer_call_fn, WeightSharedConvolutionalBoxPredictor_layer_call_fn while saving (showing 5 of 155). These functions will not be directly callable after loading.\n",
      "INFO:tensorflow:Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C208>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C2E8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C550>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], False), {}).\n",
      "I0403 11:55:32.256875 15552 def_function.py:1170] Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C208>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C2E8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C550>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], False), {}).\n",
      "INFO:tensorflow:Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A860>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A5F8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A748>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], True), {}).\n",
      "I0403 11:55:32.256875 15552 def_function.py:1170] Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A860>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A5F8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A748>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], True), {}).\n",
      "INFO:tensorflow:Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A4F28>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A46D8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A4518>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], False), {}).\n",
      "I0403 11:55:32.256875 15552 def_function.py:1170] Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A4F28>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A46D8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A4518>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], False), {}).\n",
      "INFO:tensorflow:Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095EB8>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095C88>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095D68>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], True), {}).\n",
      "I0403 11:55:32.257873 15552 def_function.py:1170] Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095EB8>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095C88>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095D68>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], True), {}).\n",
      "INFO:tensorflow:Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A4F28>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A46D8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A4518>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], False), {}).\n",
      "I0403 11:55:32.455873 15552 def_function.py:1170] Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A4F28>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A46D8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A4518>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], False), {}).\n",
      "INFO:tensorflow:Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095EB8>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095C88>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095D68>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], True), {}).\n",
      "I0403 11:55:32.455873 15552 def_function.py:1170] Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095EB8>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095C88>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095D68>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], True), {}).\n",
      "INFO:tensorflow:Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C208>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C2E8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C550>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], False), {}).\n",
      "I0403 11:55:32.455873 15552 def_function.py:1170] Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C208>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C2E8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C550>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], False), {}).\n",
      "INFO:tensorflow:Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A860>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A5F8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A748>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], True), {}).\n",
      "I0403 11:55:32.455873 15552 def_function.py:1170] Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A860>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A5F8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A748>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], True), {}).\n",
      "INFO:tensorflow:Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C208>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C2E8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C550>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], False), {}).\n",
      "I0403 11:55:32.456872 15552 def_function.py:1170] Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C208>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C2E8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C550>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], False), {}).\n",
      "INFO:tensorflow:Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A860>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A5F8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A748>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], True), {}).\n",
      "I0403 11:55:32.456872 15552 def_function.py:1170] Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A860>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A5F8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A748>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], True), {}).\n",
      "W0403 11:55:32.827284 15552 save.py:241] Found untraced functions such as WeightSharedConvolutionalBoxPredictor_layer_call_and_return_conditional_losses, WeightSharedConvolutionalBoxPredictor_layer_call_fn, WeightSharedConvolutionalBoxHead_layer_call_and_return_conditional_losses, WeightSharedConvolutionalBoxHead_layer_call_fn, WeightSharedConvolutionalBoxPredictor_layer_call_fn while saving (showing 5 of 155). These functions will not be directly callable after loading.\n",
      "INFO:tensorflow:Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C208>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C2E8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C550>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], False), {}).\n",
      "I0403 11:55:35.860894 15552 def_function.py:1170] Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C208>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C2E8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A05C550>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], False), {}).\n",
      "INFO:tensorflow:Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A860>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A5F8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A748>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], True), {}).\n",
      "I0403 11:55:35.860894 15552 def_function.py:1170] Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A860>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A5F8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A02A748>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], True), {}).\n",
      "INFO:tensorflow:Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A4F28>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A46D8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A4518>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], False), {}).\n",
      "I0403 11:55:35.860894 15552 def_function.py:1170] Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A4F28>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A46D8>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A0A4518>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], False), {}).\n",
      "INFO:tensorflow:Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095EB8>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095C88>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095D68>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], True), {}).\n",
      "I0403 11:55:35.860894 15552 def_function.py:1170] Unsupported signature for serialization: (([(<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095EB8>, TensorSpec(shape=(None, 40, 40, 32), dtype=tf.float32, name='image_features/0/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095C88>, TensorSpec(shape=(None, 20, 20, 96), dtype=tf.float32, name='image_features/1/1')), (<tensorflow.python.framework.func_graph.UnknownArgument object at 0x000002269A095D68>, TensorSpec(shape=(None, 10, 10, 1280), dtype=tf.float32, name='image_features/2/1'))], True), {}).\n",
      "INFO:tensorflow:Assets written to: Tensorflow\\workspace\\models\\my_ssd_mobnet\\tfliteexport\\saved_model\\assets\n",
      "I0403 11:55:36.624916 15552 builder_impl.py:775] Assets written to: Tensorflow\\workspace\\models\\my_ssd_mobnet\\tfliteexport\\saved_model\\assets\n"
     ]
    }
   ],
   "source": [
    "!{command}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "iJfYMbN6pfDL"
   },
   "outputs": [],
   "source": [
    "FROZEN_TFLITE_PATH = os.path.join(paths['TFLITE_PATH'], 'saved_model')\n",
    "TFLITE_MODEL = os.path.join(paths['TFLITE_PATH'], 'saved_model', 'detect.tflite')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "command = \"tflite_convert \\\n",
    "--saved_model_dir={} \\\n",
    "--output_file={} \\\n",
    "--input_shapes=1,300,300,3 \\\n",
    "--input_arrays=normalized_input_image_tensor \\\n",
    "--output_arrays='TFLite_Detection_PostProcess','TFLite_Detection_PostProcess:1','TFLite_Detection_PostProcess:2','TFLite_Detection_PostProcess:3' \\\n",
    "--inference_type=FLOAT \\\n",
    "--allow_custom_ops\".format(FROZEN_TFLITE_PATH, TFLITE_MODEL, )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "E8GwUeoFpfDL",
    "outputId": "fac43ea4-cc85-471b-a362-e994b06fd583"
   },
   "outputs": [],
   "source": [
    "print(command)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "Nbd7gqHMpfDL",
    "outputId": "7c8fe6d5-2415-4641-8548-39d425c202f7"
   },
   "outputs": [],
   "source": [
    "!{command}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "5NQqZRdA21Uc"
   },
   "source": [
    "# 13. Zip and Export Models "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "tTVTGCQp2ZJJ"
   },
   "outputs": [],
   "source": [
    "!tar -czf models.tar.gz {paths['CHECKPOINT_PATH']}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "whShhB0x3PYJ",
    "outputId": "b773201d-35c9-46a8-b893-4a76bd4d5d97"
   },
   "outputs": [],
   "source": [
    "from google.colab import drive\n",
    "drive.mount('/content/drive')"
   ]
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "name": "3. Training and Detection.ipynb",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "anprsys",
   "language": "python",
   "name": "anprsys"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}