3.5_Linear_Regression_TensorBoard.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "3.5_Linear_Regression_TensorBoard.ipynb",
      "version": "0.3.2",
      "views": {},
      "default_view": {},
      "provenance": [
        {
          "file_id": "https://github.com/DataScienceUB/DeepLearningfromScratch2018/blob/master/3.5%20Linear%20Regression%20-%20TensorBoard.ipynb",
          "timestamp": 1528921790660
        }
      ],
      "collapsed_sections": []
    },
    "kernelspec": {
      "display_name": "Python 2",
      "language": "python",
      "name": "python2"
    }
  },
  "cells": [
    {
      "metadata": {
        "id": "vyglFQhCsr7T",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "# A linear regression learning algorithm example using TensorFlow library."
      ]
    },
    {
      "metadata": {
        "id": "Zww9Rajn1-Pa",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "# How to use TensorBoard in Colab\n",
        "## Download and unzip ngrok"
      ]
    },
    {
      "metadata": {
        "id": "pSWYTvibwE-M",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          },
          "base_uri": "https://localhost:8080/",
          "height": 238
        },
        "outputId": "db947d91-cfd3-42f7-b16c-954a97262494",
        "executionInfo": {
          "status": "ok",
          "timestamp": 1528923127700,
          "user_tz": -120,
          "elapsed": 5496,
          "user": {
            "displayName": "Santi Segui",
            "photoUrl": "//lh3.googleusercontent.com/-uSLYDDb2N50/AAAAAAAAAAI/AAAAAAAAPbw/abTkAE5dwLo/s50-c-k-no/photo.jpg",
            "userId": "113371974253879421657"
          }
        }
      },
      "cell_type": "code",
      "source": [
        "! wget https://bin.equinox.io/c/4VmDzA7iaHb/ngrok-stable-linux-amd64.zip\n",
        "! unzip ngrok-stable-linux-amd64.zip"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--2018-06-13 20:52:05--  https://bin.equinox.io/c/4VmDzA7iaHb/ngrok-stable-linux-amd64.zip\r\n",
            "Resolving bin.equinox.io (bin.equinox.io)... 34.224.230.241, 34.225.24.230, 34.231.150.116, ...\n",
            "Connecting to bin.equinox.io (bin.equinox.io)|34.224.230.241|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 5363700 (5.1M) [application/octet-stream]\n",
            "Saving to: ‘ngrok-stable-linux-amd64.zip.13’\n",
            "\n",
            "ngrok-stable-linux- 100%[===================>]   5.11M  10.1MB/s    in 0.5s    \n",
            "\n",
            "2018-06-13 20:52:06 (10.1 MB/s) - ‘ngrok-stable-linux-amd64.zip.13’ saved [5363700/5363700]\n",
            "\n",
            "Archive:  ngrok-stable-linux-amd64.zip\n",
            "caution: filename not matched:  --y\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "WBOAhm082BDQ",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "##Run TensorBoard\n"
      ]
    },
    {
      "metadata": {
        "id": "XwMzI6iMwDsx",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        "LOG_DIR = './log'\n",
        "get_ipython().system_raw(\n",
        "    'tensorboard --logdir {} --host 0.0.0.0 --port 6006 &'\n",
        "    .format(LOG_DIR)\n",
        ")"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "iSEzPmdb2NCT",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "##Run ngrok"
      ]
    },
    {
      "metadata": {
        "id": "x77POjm22MrL",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        "get_ipython().system_raw('./ngrok http 6006 &')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "7ux40CVf2iiD",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "##Get URL"
      ]
    },
    {
      "metadata": {
        "id": "8DWb3Os32RQ0",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          },
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "11b4cbd8-4b6f-4a2f-f197-b23a92116632",
        "executionInfo": {
          "status": "ok",
          "timestamp": 1528923130810,
          "user_tz": -120,
          "elapsed": 1893,
          "user": {
            "displayName": "Santi Segui",
            "photoUrl": "//lh3.googleusercontent.com/-uSLYDDb2N50/AAAAAAAAAAI/AAAAAAAAPbw/abTkAE5dwLo/s50-c-k-no/photo.jpg",
            "userId": "113371974253879421657"
          }
        }
      },
      "cell_type": "code",
      "source": [
        "! curl -s http://localhost:4040/api/tunnels | python3 -c \\\n",
        "    \"import sys, json; print(json.load(sys.stdin)['tunnels'][0]['public_url'])\""
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "http://cd6f37bf.ngrok.io\r\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "SSnjNtXw2RaU",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "# Train the model with tensorboard summaries"
      ]
    },
    {
      "metadata": {
        "id": "hUU81dbBsr7V",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "Linear regression model is one of the simplest regression models. It assumes linear relationship between X and Y. The output equation is defined as follows:\n",
        "$$\\hat{y} = WX + b$$\n"
      ]
    },
    {
      "metadata": {
        "id": "QG19FUtusr7W",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import pandas as pd\n",
        "%matplotlib inline"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "WlAioeH7sr7Y",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        "#  Training Parameters\n",
        "learning_rate = 0.001\n",
        "training_epochs = 10000\n",
        "display_step = 200"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "pq4RH82hsr7b",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          },
          "base_uri": "https://localhost:8080/",
          "height": 299
        },
        "outputId": "7a0f08d9-a259-4eba-ff94-d79b20e7879a",
        "executionInfo": {
          "status": "ok",
          "timestamp": 1528923133423,
          "user_tz": -120,
          "elapsed": 889,
          "user": {
            "displayName": "Santi Segui",
            "photoUrl": "//lh3.googleusercontent.com/-uSLYDDb2N50/AAAAAAAAAAI/AAAAAAAAPbw/abTkAE5dwLo/s50-c-k-no/photo.jpg",
            "userId": "113371974253879421657"
          }
        }
      },
      "cell_type": "code",
      "source": [
        "# Load data. Advertising dataset from \"An Introduction to Statistical Learning\",\n",
        "# textbook by Gareth James, Robert Tibshirani, and Trevor Hastie \n",
        "import numpy as np\n",
        "data = pd.read_csv('https://raw.githubusercontent.com/DataScienceUB/DeepLearningfromScratch2017/master/dataset/Advertising.csv',index_col=0, delimiter=',')\n",
        "train_X = data[['TV']].values \n",
        "\n",
        "train_Y = data.Sales.values \n",
        "train_Y = train_Y[:,np.newaxis]\n",
        "\n",
        "n_samples = train_X.shape[0]\n",
        "print n_samples\n",
        "print train_X.shape, train_Y.shape\n",
        "plt.plot(train_X, train_Y, 'ro', label='Original data')\n",
        "plt.show()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "200\n",
            "(200, 1) (200, 1)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAW8AAAD4CAYAAAAjKGdbAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nrkJggg==\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fb8f6b70850>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "IgJIt0VJsr7e",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        "import tensorflow as tf\n",
        "# Define tf Graph Inputs\n",
        "X = tf.placeholder(\"float\",[None,1])\n",
        "y = tf.placeholder(\"float\",[None,1])\n",
        "\n",
        "# Create Model variables \n",
        "# Set model weights\n",
        "W = tf.Variable(np.random.randn(), name=\"weight\")\n",
        "b = tf.Variable(np.random.randn(), name=\"bias\")\n",
        "\n",
        "# Construct a linear model\n",
        "y_pred = tf.add(tf.multiply(X, W), b)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "M9BKUh6tsr7g",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        "# Minimize the squared errors\n",
        "cost = tf.reduce_sum(tf.pow(y_pred-y,2))/(2*n_samples) #L2 loss\n",
        "\n",
        "# Define the optimizer\n",
        "optimizer = tf.train.AdamOptimizer(learning_rate).minimize(cost) #Gradient descent"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "QluV2g93sr7i",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        "# Add summary ops to collect data\n",
        "w_hist = tf.summary.histogram(\"weights\", W)\n",
        "b_hist = tf.summary.histogram(\"biases\", b)\n",
        "y_hist = tf.summary.histogram(\"y\", y_pred)\n",
        "\n",
        "cost_summary = tf.summary.scalar(\"cost\", cost)\n",
        "\n",
        "# Merge all the summaries and write them out to /tmp/linear_regression\n",
        "merged = tf.summary.merge_all()\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "mJ2q-_xKsr7m",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          },
          "base_uri": "https://localhost:8080/",
          "height": 1149
        },
        "outputId": "55f4d206-5a79-427f-c966-f4e854e19030",
        "executionInfo": {
          "status": "ok",
          "timestamp": 1528923147952,
          "user_tz": -120,
          "elapsed": 12231,
          "user": {
            "displayName": "Santi Segui",
            "photoUrl": "//lh3.googleusercontent.com/-uSLYDDb2N50/AAAAAAAAAAI/AAAAAAAAPbw/abTkAE5dwLo/s50-c-k-no/photo.jpg",
            "userId": "113371974253879421657"
          }
        }
      },
      "cell_type": "code",
      "source": [
        "import time\n",
        "\n",
        "# Initializing the variables\n",
        "init = tf.global_variables_initializer()\n",
        "# Launch the graph\n",
        "with tf.Session() as sess:\n",
        "    summary_writer = tf.summary.FileWriter(LOG_DIR, graph=tf.get_default_graph())\n",
        "    \n",
        "    sess.run(init)\n",
        "\n",
        "    # Fit all training data\n",
        "    for epoch in range(training_epochs):\n",
        "        _,_,summary = sess.run([optimizer, cost, merged], feed_dict={X: train_X, y: train_Y}) \n",
        "        summary_writer.add_summary(summary, epoch)\n",
        "        #Display logs per epoch step\n",
        "        if epoch % display_step == 0:\n",
        "            print \"Epoch:\", '%04d' % (epoch+1), \"cost=\", \\\n",
        "                \"{:.9f}\".format(sess.run(cost, feed_dict={X: train_X, y:train_Y})), \\\n",
        "                \"W=\", sess.run(W), \"b=\", sess.run(b)\n",
        "\n",
        "    print \"Optimization Finished!\"\n",
        "    print \"cost=\", sess.run(cost, feed_dict={X: train_X, y: train_Y}), \\\n",
        "          \"W=\", sess.run(W), \"b=\", sess.run(b)\n",
        "\n",
        "    #Graphic display\n",
        "    plt.plot(train_X, train_Y, 'ro', label='Original data')\n",
        "    plt.plot(train_X, sess.run(W) * train_X + sess.run(b), label='Fitted line')\n",
        "    plt.legend()\n",
        "    plt.show()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch: 0001 cost= 8810.703125000 W= -0.6894207 b= -1.3411077\n",
            "Epoch: 0201 cost= 5048.647460938 W= -0.500618 b= -1.1520994\n",
            "Epoch: 0401 cost= 2661.365966797 W= -0.3393883 b= -0.9900558\n",
            "Epoch: 0601 cost= 1271.116333008 W= -0.20717394 b= -0.85618573\n",
            "Epoch: 0801 cost= 542.542724609 W= -0.10420147 b= -0.75045925\n",
            "Epoch: 1001 cost= 206.509521484 W= -0.029021127 b= -0.6711899\n",
            "Epoch: 1201 cost= 73.067291260 W= 0.02177902 b= -0.6148045\n",
            "Epoch: 1401 cost= 28.344144821 W= 0.053186923 b= -0.57625353\n",
            "Epoch: 1601 cost= 15.910450935 W= 0.07077963 b= -0.54997903\n",
            "Epoch: 1801 cost= 13.080209732 W= 0.07962239 b= -0.53095925\n",
            "Epoch: 2001 cost= 12.548313141 W= 0.08356252 b= -0.5153671\n",
            "Epoch: 2201 cost= 12.450679779 W= 0.08508392 b= -0.5006927\n",
            "Epoch: 2401 cost= 12.414784431 W= 0.085560046 b= -0.48548976\n",
            "Epoch: 2601 cost= 12.383067131 W= 0.08564154 b= -0.46899694\n",
            "Epoch: 2801 cost= 12.348598480 W= 0.08559098 b= -0.45081452\n",
            "Epoch: 3001 cost= 12.310529709 W= 0.08549743 b= -0.4306954\n",
            "Epoch: 3201 cost= 12.268524170 W= 0.0853857 b= -0.4084398\n",
            "Epoch: 3401 cost= 12.222255707 W= 0.085260734 b= -0.38385135\n",
            "Epoch: 3601 cost= 12.171388626 W= 0.08512266 b= -0.35672334\n",
            "Epoch: 3801 cost= 12.115559578 W= 0.08497048 b= -0.32683408\n",
            "Epoch: 4001 cost= 12.054394722 W= 0.084803045 b= -0.293947\n",
            "Epoch: 4201 cost= 11.987501144 W= 0.084619045 b= -0.25781152\n",
            "Epoch: 4401 cost= 11.914486885 W= 0.08441714 b= -0.21816348\n",
            "Epoch: 4601 cost= 11.834954262 W= 0.084195934 b= -0.17472818\n",
            "Epoch: 4801 cost= 11.748515129 W= 0.08395398 b= -0.12722254\n",
            "Epoch: 5001 cost= 11.654802322 W= 0.083689794 b= -0.07535931\n",
            "Epoch: 5201 cost= 11.553468704 W= 0.08340193 b= -0.01885148\n",
            "Epoch: 5401 cost= 11.444226265 W= 0.08308894 b= 0.042581037\n",
            "Epoch: 5601 cost= 11.326829910 W= 0.08274947 b= 0.109204456\n",
            "Epoch: 5801 cost= 11.201124191 W= 0.082382284 b= 0.18126269\n",
            "Epoch: 6001 cost= 11.067038536 W= 0.0819863 b= 0.25896809\n",
            "Epoch: 6201 cost= 10.924619675 W= 0.08156062 b= 0.34249195\n",
            "Epoch: 6401 cost= 10.774034500 W= 0.08110465 b= 0.4319551\n",
            "Epoch: 6601 cost= 10.615583420 W= 0.080618076 b= 0.527419\n",
            "Epoch: 6801 cost= 10.449711800 W= 0.08010094 b= 0.6288793\n",
            "Epoch: 7001 cost= 10.276999474 W= 0.079553634 b= 0.73625994\n",
            "Epoch: 7201 cost= 10.098161697 W= 0.078976914 b= 0.8494137\n",
            "Epoch: 7401 cost= 9.914031982 W= 0.07837191 b= 0.9681227\n",
            "Epoch: 7601 cost= 9.725531578 W= 0.07774009 b= 1.0921048\n",
            "Epoch: 7801 cost= 9.533662796 W= 0.077083185 b= 1.2210195\n",
            "Epoch: 8001 cost= 9.339451790 W= 0.07640317 b= 1.3544872\n",
            "Epoch: 8201 cost= 9.143948555 W= 0.075702175 b= 1.49209\n",
            "Epoch: 8401 cost= 8.948181152 W= 0.07498245 b= 1.6333916\n",
            "Epoch: 8601 cost= 8.753134727 W= 0.07424624 b= 1.7779493\n",
            "Epoch: 8801 cost= 8.559740067 W= 0.073495835 b= 1.925323\n",
            "Epoch: 9001 cost= 8.368845940 W= 0.07273338 b= 2.0750847\n",
            "Epoch: 9201 cost= 8.181225777 W= 0.071960986 b= 2.226824\n",
            "Epoch: 9401 cost= 7.997556210 W= 0.07118061 b= 2.380159\n",
            "Epoch: 9601 cost= 7.818436146 W= 0.07039408 b= 2.5347285\n",
            "Epoch: 9801 cost= 7.644371986 W= 0.069603056 b= 2.6902025\n",
            "Optimization Finished!\n",
            "cost= 7.4766245 W= 0.06881309 b= 2.8454936\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAW8AAAD4CAYAAAAjKGdbAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nEFLeBEEQJoSUN0EQhAkh5U0QBGFC/h/AC2RVrcr5QAAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fb8f5ce6810>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "drAb0Gq7sr7p",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        "#!nohup tensorboard --logdir ."
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "yV3z94iou4NR",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "T0kc5twswsOO",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          },
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "4488928a-90cd-4cc8-9a9f-62c8fceb9cb8",
        "executionInfo": {
          "status": "ok",
          "timestamp": 1528923150974,
          "user_tz": -120,
          "elapsed": 1685,
          "user": {
            "displayName": "Santi Segui",
            "photoUrl": "//lh3.googleusercontent.com/-uSLYDDb2N50/AAAAAAAAAAI/AAAAAAAAPbw/abTkAE5dwLo/s50-c-k-no/photo.jpg",
            "userId": "113371974253879421657"
          }
        }
      },
      "cell_type": "code",
      "source": [
        "ls log"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "events.out.tfevents.1528923135.a85a5d1dfbb6\r\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "IPQNIZjwxPdW",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          },
          "base_uri": "https://localhost:8080/",
          "height": 68
        },
        "outputId": "03dc8082-37fb-4ddd-df50-a27923419005",
        "executionInfo": {
          "status": "ok",
          "timestamp": 1528923152735,
          "user_tz": -120,
          "elapsed": 1731,
          "user": {
            "displayName": "Santi Segui",
            "photoUrl": "//lh3.googleusercontent.com/-uSLYDDb2N50/AAAAAAAAAAI/AAAAAAAAPbw/abTkAE5dwLo/s50-c-k-no/photo.jpg",
            "userId": "113371974253879421657"
          }
        }
      },
      "cell_type": "code",
      "source": [
        "!ps aux | grep tensorboard"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "root      2382 32.4  1.4 877672 196460 ?       Sl   20:52   0:08 /usr/bin/python2 /usr/local/bin/tensorboard --logdir ./log --host 0.0.0.0 --port 6006\r\n",
            "root      2497 54.5  0.0  33960  5096 pts/1    Ss+  20:52   0:01 /bin/sh -c ps aux | grep tensorboard\r\n",
            "root      2499  0.0  0.0  38200  5608 pts/1    S+   20:52   0:00 grep tensorboard\r\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "Q-usEFtb6o9R",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        "#!kill 408"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "ENs8UQpa6xnI",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        "#!rm -rf log"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "hqb0y5Jd6zgm",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}