keras_to_caffe2_demo.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import packages and load MNIST dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X_train shape (60000, 28, 28)\n",
      "y_train shape (60000,)\n"
     ]
    }
   ],
   "source": [
    "### from keras.models import Sequential\n",
    "from keras.layers.core import Dense, Dropout, Activation\n",
    "from keras.utils import np_utils\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "#import mnist data set\n",
    "from keras.datasets import mnist\n",
    "\n",
    "# the data, shuffled and split between tran and test sets\n",
    "(X_train, Y_train), (X_test, Y_test) = mnist.load_data()\n",
    "print(\"X_train shape\", X_train.shape)\n",
    "print(\"y_train shape\", Y_train.shape)\n",
    "\n",
    "IMAGE_H = X_train.shape[1]\n",
    "IMAGE_W = X_train.shape[2]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Show samples of the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 12 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8, 8))\n",
    "for i in range(12):\n",
    "    plt.subplot(3, 4, i+1)\n",
    "    plt.imshow(X_train[i], cmap='gray')\n",
    "    plt.title(\"Class {}\".format(Y_train[i]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create keras model to train MNIST data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "data (InputLayer)            (None, 28, 28, 1)         0         \n",
      "_________________________________________________________________\n",
      "conv1 (Conv2D)               (None, 24, 24, 32)        832       \n",
      "_________________________________________________________________\n",
      "pool1 (MaxPooling2D)         (None, 12, 12, 32)        0         \n",
      "_________________________________________________________________\n",
      "conv2 (Conv2D)               (None, 10, 10, 64)        18496     \n",
      "_________________________________________________________________\n",
      "pool2 (MaxPooling2D)         (None, 5, 5, 64)          0         \n",
      "_________________________________________________________________\n",
      "flatten (Flatten)            (None, 1600)              0         \n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            (None, 1600)              0         \n",
      "_________________________________________________________________\n",
      "dense1 (Dense)               (None, 256)               409856    \n",
      "_________________________________________________________________\n",
      "output (Dense)               (None, 10)                2570      \n",
      "=================================================================\n",
      "Total params: 431,754\n",
      "Trainable params: 431,754\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "from keras import backend as K\n",
    "from keras.models import Sequential, Model\n",
    "from keras.layers import Dense, Dropout, Flatten, Input, Reshape, AveragePooling2D\n",
    "from keras.layers import Conv2D, MaxPooling2D, BatchNormalization, concatenate\n",
    "from keras.layers import Softmax\n",
    "\n",
    "#d_format = 'channels_first'\n",
    "d_format='channels_last'\n",
    "\n",
    "if d_format == 'channels_last':\n",
    "    input_image = Input(shape=(IMAGE_H, IMAGE_W, 1), name='data')\n",
    "else:\n",
    "    input_image = Input(shape=(1, IMAGE_H, IMAGE_W), name='data')\n",
    "\n",
    "conv1 = Conv2D(32, (5, 5), strides=(1,1), name='conv1', data_format=d_format)(input_image)\n",
    "pool1 = MaxPooling2D(pool_size=(2, 2), name='pool1', data_format=d_format)(conv1)\n",
    "conv2 = Conv2D(64, (3,3), strides=(1,1), name='conv2', data_format=d_format)(pool1)\n",
    "pool2 = MaxPooling2D(pool_size=(2, 2), name='pool2', data_format=d_format)(conv2)\n",
    "\n",
    "x = Flatten(name='flatten')(pool2)\n",
    "x = Dropout(0.5, name='dropout')(x);\n",
    "\n",
    "fc1 = Dense(256, name = 'dense1', activation='relu')(x)\n",
    "out = Dense(10, name='output', activation='softmax')(fc1)\n",
    "\n",
    "model = Model(input_image, out)\n",
    "\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preprocess the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training input shape (60000, 28, 28, 1)\n",
      "training output shape  (60000, 10)\n",
      "Testing input shape (10000, 28, 28, 1)\n",
      "testing output shape  (10000, 10)\n"
     ]
    }
   ],
   "source": [
    "x_train = X_train.reshape(X_train.shape[0], X_train.shape[1], X_train.shape[2], 1)\n",
    "x_test = X_test.reshape(X_test.shape[0], X_test.shape[1], X_test.shape[2], 1)\n",
    "\n",
    "x_train = x_train.astype('float32')\n",
    "x_test = x_test.astype('float32')\n",
    "x_train /= 255\n",
    "x_test /= 255\n",
    "\n",
    "n_classes = 10\n",
    "\n",
    "y_train = np_utils.to_categorical(Y_train, n_classes)\n",
    "y_test = np_utils.to_categorical(Y_test, n_classes)\n",
    "\n",
    "print(\"Training input shape\", x_train.shape)\n",
    "print('training output shape ', y_train.shape)\n",
    "print(\"Testing input shape\", x_test.shape)\n",
    "print('testing output shape ', y_test.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Compile the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train the model\n",
    "Note: Depending on your computer's resources, the training step may take a while. My computer has NVIDIA Quadro M1000M and the training takes a little over one minute."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 60000 samples, validate on 10000 samples\n",
      "Epoch 1/8\n",
      "60000/60000 [==============================] - 11s 188us/step - loss: 0.2095 - acc: 0.9366 - val_loss: 0.0531 - val_acc: 0.9838\n",
      "Epoch 2/8\n",
      "60000/60000 [==============================] - 9s 150us/step - loss: 0.0673 - acc: 0.9793 - val_loss: 0.0328 - val_acc: 0.9880\n",
      "Epoch 3/8\n",
      "60000/60000 [==============================] - 9s 151us/step - loss: 0.0483 - acc: 0.9845 - val_loss: 0.0313 - val_acc: 0.9903\n",
      "Epoch 4/8\n",
      "60000/60000 [==============================] - 9s 151us/step - loss: 0.0409 - acc: 0.9869 - val_loss: 0.0257 - val_acc: 0.9924\n",
      "Epoch 5/8\n",
      "60000/60000 [==============================] - 9s 150us/step - loss: 0.0344 - acc: 0.9890 - val_loss: 0.0307 - val_acc: 0.9905\n",
      "Epoch 6/8\n",
      "60000/60000 [==============================] - 9s 148us/step - loss: 0.0283 - acc: 0.9909 - val_loss: 0.0284 - val_acc: 0.9916\n",
      "Epoch 7/8\n",
      "60000/60000 [==============================] - 9s 149us/step - loss: 0.0268 - acc: 0.9918 - val_loss: 0.0285 - val_acc: 0.9910\n",
      "Epoch 8/8\n",
      "60000/60000 [==============================] - 9s 149us/step - loss: 0.0254 - acc: 0.9914 - val_loss: 0.0333 - val_acc: 0.9906\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x20195f88f60>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(x_train, y_train,\n",
    "          batch_size=128, epochs=8,\n",
    "          verbose=1,\n",
    "          validation_data=(x_test, y_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Predict outpus for all test samples"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# The predict_classes function outputs the highest probability class\n",
    "# according to the trained classifier for each input example.\n",
    "pred_test = model.predict(x_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Show output predictions for some randomly selected test inputs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 12 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "rind = np.arange(0, x_test.shape[0])\n",
    "np.random.shuffle(rind)\n",
    "\n",
    "plt.figure(figsize=(8, 8))\n",
    "plt.subplots_adjust(left=1,  right=2)\n",
    "\n",
    "for i in range(0, 12):\n",
    "    ind  = rind[i]\n",
    "    plt.subplot(3,4,i+1)\n",
    "    \n",
    "    plt.imshow(X_test[ind].reshape(28,28), cmap='gray', interpolation='none')\n",
    "    pred_class = np.argmax(pred_test[ind])\n",
    "    plt.title(\"Predicted {}, Class {}\".format(pred_class, Y_test[ind]))\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "outFileName = './mnist_keras_model.h5'\n",
    "model.save(outFileName)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Convert keras model to caffe2 model and save it to file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras2caffe2 import keras2caffe2\n",
    "\n",
    "init_net = './caffe2_init_net.pb'\n",
    "predict_net = './caffe2_predict_net.pb'\n",
    "\n",
    "train_net, test_net =  keras2caffe2(model, IMAGE_W, IMAGE_H,\n",
    "                                    init_net, predict_net, \n",
    "                                    use_cudnn=True, \n",
    "                                    keras_channel_last= d_format == 'channels_last'\n",
    "                                   )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Clear keras memory so that caffe2 can use all available GPU memory"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras import backend as K \n",
    "\n",
    "sess = K.get_session()\n",
    "K.clear_session()\n",
    "sess.close()\n",
    "\n",
    "#use numba library to release cuda memory\n",
    "from numba import cuda\n",
    "cuda.select_device(0)\n",
    "cuda.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Loade saved caffe2 model from file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "from caffe2.python import workspace, core, model_helper, brew, optimizer, utils\n",
    "from caffe2.proto import caffe2_pb2\n",
    "\n",
    "def load_net(INIT_NET, PREDICT_NET, device_opts):\n",
    "\n",
    "    init_def = caffe2_pb2.NetDef()\n",
    "    with open(INIT_NET, 'rb') as f:\n",
    "        init_def.ParseFromString(f.read())\n",
    "        init_def.device_option.CopyFrom(device_opts)\n",
    "        workspace.RunNetOnce(init_def.SerializeToString())\n",
    "    \n",
    "    net_def = caffe2_pb2.NetDef()\n",
    "    with open(PREDICT_NET, 'rb') as f:\n",
    "        net_def.ParseFromString(f.read())\n",
    "        net_def.device_option.CopyFrom(device_opts)\n",
    "        workspace.CreateNet(net_def.SerializeToString(), overwrite=True)\n",
    "        \n",
    "    return (init_def, net_def)\n",
    "    \n",
    "\n",
    "init_net = './caffe2_init_net.pb'\n",
    "predict_net = './caffe2_predict_net.pb'\n",
    "\n",
    "device_opts = core.DeviceOption(caffe2_pb2.CUDA, 0) \n",
    "\n",
    "init_def, net_def = load_net(init_net, predict_net, device_opts)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Display caffe2 model "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 13,
     "metadata": {
      "image/png": {
       "height": 1200,
       "width": 1800
      }
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from caffe2.python import net_drawer\n",
    "from IPython import display\n",
    "from caffe2.python import workspace, core, model_helper, brew, optimizer, utils\n",
    "from caffe2.proto import caffe2_pb2\n",
    "\n",
    "graph = net_drawer.GetPydotGraph(net_def, rankdir=\"LR\")\n",
    "display.Image(graph.create_png(), width=1800, height=1200)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Predict the randomly selected test samples using caffe2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 12 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "\n",
    "plt.figure(figsize=(8, 8))\n",
    "plt.subplots_adjust(left=1,  right=2)\n",
    "\n",
    "device_opts = core.DeviceOption(caffe2_pb2.CUDA, 0) \n",
    "\n",
    "for i in range(0, 12):\n",
    "    ind  = rind[i]\n",
    "    plt.subplot(3,4,i+1)\n",
    "    \n",
    "    test_img = X_test[ind].reshape(IMAGE_H, IMAGE_W)\n",
    "    plt.imshow(test_img, cmap='gray')\n",
    "    \n",
    "    test_img_cfirst = x_test[ind].reshape(1, 1, IMAGE_H, IMAGE_W)\n",
    "    workspace.FeedBlob(\"data\", test_img_cfirst, device_option=device_opts)\n",
    "    workspace.RunNet(net_def.name, 1)\n",
    "    \n",
    "    caffe2_pred_test = workspace.FetchBlob(\"softmax\"),   \n",
    "    pred_class = np.argmax(caffe2_pred_test)\n",
    "    \n",
    "    plt.title(\"Predicted {}, Class {}\".format(pred_class, Y_test[ind]))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Caffe2's outputs should match outputs from keras model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}