diff --git a/GTSRB/03-Tracking-and-visualizing.ipynb b/GTSRB/03-Tracking-and-visualizing.ipynb index e1eb8572d944182bdb5d698e3fdb342c1e0add85..bca459c4b9600d0f53ef4511557cd7cfe43cd243 100644 --- a/GTSRB/03-Tracking-and-visualizing.ipynb +++ b/GTSRB/03-Tracking-and-visualizing.ipynb @@ -331,7 +331,6 @@ "metadata": {}, "outputs": [], "source": [ - "reload(ooo)\n", "y_pred = model.predict_classes(x_test)\n", "conf_mat = confusion_matrix(y_test,y_pred, normalize=\"true\", labels=range(43))\n", "\n", diff --git a/GTSRB/04-Data-augmentation.ipynb b/GTSRB/04-Data-augmentation.ipynb index ebc18d8461049bcc86ae430927babcf280e7ee17..870955df5a7f8f2201be5899a0cae56af82e3a69 100644 --- a/GTSRB/04-Data-augmentation.ipynb +++ b/GTSRB/04-Data-augmentation.ipynb @@ -9,14 +9,10 @@ "---\n", "Introduction au Deep Learning (IDLE) - S. Aria, E. Maldonado, JL. Parouty - CNRS/SARI/DEVLOG - 2020\n", "\n", - "## Episode 3 : Tracking, visualizing and save models\n", + "## Episode 4 : Data augmentation\n", "\n", "Our main steps:\n", - " - Monitoring and understanding our model training\n", - " - Analyze the results \n", - " - Improving our model\n", - " - Add recovery points\n", - "\n", + " - Increase and improve the learning dataset\n", "\n", "## 1/ Import and init" ] @@ -32,7 +28,7 @@ "text": [ "IDLE 2020 - Practical Work Module\n", " Version : 0.1.4\n", - " Run time : Saturday 11 January 2020, 00:09:42\n", + " Run time : Monday 13 January 2020, 21:15:42\n", " Matplotlib style : idle/talk.mplstyle\n", " TensorFlow version : 2.0.0\n", " Keras version : 2.2.4-tf\n" @@ -46,7 +42,11 @@ "\n", "import numpy as np\n", "import h5py\n", + "\n", + "from sklearn.metrics import confusion_matrix\n", + "\n", "import matplotlib.pyplot as plt\n", + "import seaborn as sn\n", "import os, time, random\n", "\n", "import idle.pwk as ooo\n", @@ -59,7 +59,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## 2/ Reload dataset\n", + "## 2/ Dataset loader\n", "Dataset is one of the saved dataset: RGB25, RGB35, L25, L35, etc. \n", "First of all, we're going to use a smart dataset : **set-24x24-L** \n", "(with a GPU, it only takes 35'' compared to more than 5' with a CPU !)" @@ -67,258 +67,578 @@ }, { "cell_type": "code", - "execution_count": 2, + "execution_count": 11, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "Dataset loaded (228.8 Mo)\n", + "Dataset \"set-24x24-L-LHE\" is loaded. (228.8 Mo)\n", "\n", - "CPU times: user 0 ns, sys: 297 ms, total: 297 ms\n", - "Wall time: 294 ms\n" + "CPU times: user 8 ms, sys: 132 ms, total: 140 ms\n", + "Wall time: 1.32 s\n" ] } ], "source": [ "%%time\n", "\n", - "dataset ='set-24x24-L'\n", + "def read_dataset(name):\n", + " '''Reads h5 dataset from ./data\n", "\n", - "# ---- Read dataset\n", + " Arguments: dataset name, without .h5\n", + " Returns: x_train,y_train,x_test,y_test data'''\n", + " # ---- Read dataset\n", + " filename='./data/'+name+'.h5'\n", + " with h5py.File(filename) as f:\n", + " x_train = f['x_train'][:]\n", + " y_train = f['y_train'][:]\n", + " x_test = f['x_test'][:]\n", + " y_test = f['y_test'][:]\n", + "\n", + " # ---- done\n", + " print('Dataset \"{}\" is loaded. ({:.1f} Mo)\\n'.format(name,os.path.getsize(filename)/(1024*1024)))\n", + " return x_train,y_train,x_test,y_test" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## 3/ Models\n", + "We will now build a model and train it...\n", + "\n", + "This is my model ;-) " + ] + }, + { + "cell_type": "code", + "execution_count": 37, + "metadata": {}, + "outputs": [], + "source": [ + "# A basic model\n", "#\n", - "filename='./data/'+dataset+'.h5'\n", - "with h5py.File(filename) as f:\n", - " x_train = f['x_train'][:]\n", - " y_train = f['y_train'][:]\n", - " x_test = f['x_test'][:]\n", - " y_test = f['y_test'][:]\n", - "\n", - "# ---- Dataset shape\n", + "def get_model_v1(lx,ly,lz):\n", + " \n", + " model = keras.models.Sequential()\n", + " \n", + " model.add( keras.layers.Conv2D(96, (3,3), activation='relu', input_shape=(lx,ly,lz)))\n", + " model.add( keras.layers.MaxPooling2D((2, 2)))\n", + " model.add( keras.layers.Dropout(0.2))\n", + "\n", + " model.add( keras.layers.Conv2D(192, (3, 3), activation='relu'))\n", + " model.add( keras.layers.MaxPooling2D((2, 2)))\n", + " model.add( keras.layers.Dropout(0.2))\n", + "\n", + " model.add( keras.layers.Flatten()) \n", + " model.add( keras.layers.Dense(1500, activation='relu'))\n", + " model.add( keras.layers.Dropout(0.5))\n", + "\n", + " model.add( keras.layers.Dense(43, activation='softmax'))\n", + " return model\n", + "\n", + "# A more sophisticated model\n", "#\n", - "(n,lx,ly,lz) = x_train.shape\n", - "data_shape = (lx, ly, lz)\n", + "def get_model_v2(lx,ly,lz):\n", + " model = keras.models.Sequential()\n", + "\n", + " model.add( keras.layers.Conv2D(64, (3, 3), padding='same', input_shape=(lx,ly,lz), activation='relu'))\n", + " model.add( keras.layers.Conv2D(64, (3, 3), activation='relu'))\n", + " model.add( keras.layers.MaxPooling2D(pool_size=(2, 2)))\n", + " model.add( keras.layers.Dropout(0.2))\n", + "\n", + " model.add( keras.layers.Conv2D(128, (3, 3), padding='same', activation='relu'))\n", + " model.add( keras.layers.Conv2D(128, (3, 3), activation='relu'))\n", + " model.add( keras.layers.MaxPooling2D(pool_size=(2, 2)))\n", + " model.add( keras.layers.Dropout(0.2))\n", "\n", - "# ---- done\n", - "print('Dataset loaded ({:.1f} Mo)\\n'.format(os.path.getsize(filename)/(1024*1024)))" + " model.add( keras.layers.Conv2D(256, (3, 3), padding='same',activation='relu'))\n", + " model.add( keras.layers.Conv2D(256, (3, 3), activation='relu'))\n", + " model.add( keras.layers.MaxPooling2D(pool_size=(2, 2)))\n", + " model.add( keras.layers.Dropout(0.2))\n", + "\n", + " model.add( keras.layers.Flatten())\n", + " model.add( keras.layers.Dense(512, activation='relu'))\n", + " model.add( keras.layers.Dropout(0.5))\n", + " model.add( keras.layers.Dense(43, activation='softmax'))\n", + " return model\n", + "\n", + "# My sphisticated model, but small and fast\n", + "#\n", + "def get_model_v3(lx,ly,lz):\n", + " model = keras.models.Sequential()\n", + " model.add( keras.layers.Conv2D(32, (3,3), activation='relu', input_shape=(lx,ly,lz)))\n", + " model.add( keras.layers.MaxPooling2D((2, 2)))\n", + " model.add( keras.layers.Dropout(0.5))\n", + "\n", + " model.add( keras.layers.Conv2D(64, (3, 3), activation='relu'))\n", + " model.add( keras.layers.MaxPooling2D((2, 2)))\n", + " model.add( keras.layers.Dropout(0.5))\n", + "\n", + " model.add( keras.layers.Conv2D(128, (3, 3), activation='relu'))\n", + " model.add( keras.layers.MaxPooling2D((2, 2)))\n", + " model.add( keras.layers.Dropout(0.5))\n", + "\n", + " model.add( keras.layers.Conv2D(256, (3, 3), activation='relu'))\n", + " model.add( keras.layers.MaxPooling2D((2, 2)))\n", + " model.add( keras.layers.Dropout(0.5))\n", + "\n", + " model.add( keras.layers.Flatten()) \n", + " model.add( keras.layers.Dense(1152, activation='relu'))\n", + " model.add( keras.layers.Dropout(0.5))\n", + "\n", + " model.add( keras.layers.Dense(43, activation='softmax'))\n", + " return model" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "## 3/ Have a look to the dataset\n", - "Note: Data must be reshape for matplotlib" + "## 4/ Callbacks \n", + "We prepare 2 kind callbacks : TensorBoard and Model backup" ] }, { "cell_type": "code", - "execution_count": 4, + "execution_count": 31, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "x_train : (39209, 24, 24, 1)\n", - "y_train : (39209,)\n", - "x_test : (12630, 24, 24, 1)\n", - "y_test : (12630,)\n" + "total 8\n", + "drwxr-x--- 3 paroutyj l-simap 4096 Jan 13 21:28 logs\n", + "drwxr-x--- 2 paroutyj l-simap 4096 Jan 13 21:33 models\n" ] - }, - { - "data": { - "image/png": "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\n", - "text/plain": [ - "<Figure size 864x291.6 with 36 Axes>" - ] - }, - "metadata": {}, - "output_type": "display_data" } ], "source": [ - "print(\"x_train : \", x_train.shape)\n", - "print(\"y_train : \", y_train.shape)\n", - "print(\"x_test : \", x_test.shape)\n", - "print(\"y_test : \", y_test.shape)\n", + "%%bash\n", + "# To clean old logs and saved model, run this cell\n", + "#\n", + "#/bin/rm -r ./run/logs 2>/dev/null\n", + "#/bin/rm -r ./run/models 2>/dev/null\n", + "/bin/ls -l ./run 2>/dev/null" + ] + }, + { + "cell_type": "code", + "execution_count": 32, + "metadata": {}, + "outputs": [], + "source": [ + "ooo.mkdir('./run/models')\n", + "ooo.mkdir('./run/logs')\n", + "\n", + "# ---- Callback tensorboard\n", + "log_dir = \"./run/logs/tb_\" + ooo.tag_now()\n", + "tensorboard_callback = tf.keras.callbacks.TensorBoard(log_dir=log_dir, histogram_freq=1)\n", "\n", - "ooo.plot_images(x_train, y_train, range(36), columns=12, x_size=1, y_size=1)\n" + "# ---- Callback ModelCheckpoint - Save best model\n", + "save_dir = \"./run/models/best-model.h5\"\n", + "bestmodel_callback = tf.keras.callbacks.ModelCheckpoint(filepath=save_dir, verbose=0, monitor='accuracy', save_best_only=True)\n", + "\n", + "# ---- Callback ModelCheckpoint - Save model each epochs\n", + "save_dir = \"./run/models/model-{epoch:04d}.h5\"\n", + "savemodel_callback = tf.keras.callbacks.ModelCheckpoint(filepath=save_dir, verbose=0, save_freq=2000*5)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "## 4/ Create model\n", - "Some hyperparameters :" + "## 5/ Load and prepare dataset\n", + "### 5.1/ Load" ] }, { "cell_type": "code", - "execution_count": 10, + "execution_count": 33, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Dataset \"set-48x48-L-LHE\" is loaded. (913.9 Mo)\n", + "\n" + ] + } + ], "source": [ - "batch_size = 64\n", - "num_classes = 43\n", - "epochs = 8" + "x_train,y_train,x_test,y_test = read_dataset('set-48x48-L-LHE')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "My models :" + "### 5.2/ Data augmentation" ] }, { "cell_type": "code", - "execution_count": 32, + "execution_count": 34, "metadata": {}, "outputs": [], "source": [ - "\n", - "def get_model_v1():\n", - " model = keras.models.Sequential()\n", - " model.add( keras.layers.Conv2D(96, (3,3), activation='relu', input_shape=(lx,ly,lz)))\n", - " model.add( keras.layers.MaxPooling2D((2, 2)))\n", - " model.add( keras.layers.Conv2D(192, (3, 3), activation='relu'))\n", - " model.add( keras.layers.MaxPooling2D((2, 2)))\n", - " model.add( keras.layers.Flatten()) \n", - " model.add( keras.layers.Dense(500, activation='relu'))\n", - " model.add( keras.layers.Dense(500, activation='relu'))\n", - " model.add( keras.layers.Dense(43, activation='softmax'))\n", - " return model\n" + "datagen = keras.preprocessing.image.ImageDataGenerator(featurewise_center=False,\n", + " featurewise_std_normalization=False,\n", + " width_shift_range=0.1,\n", + " height_shift_range=0.1,\n", + " zoom_range=0.2,\n", + " shear_range=0.1,\n", + " rotation_range=10.)\n", + "datagen.fit(x_train)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## 5/ Train the model\n", + "**Get the shape of my data :**" ] }, { "cell_type": "code", - "execution_count": 33, + "execution_count": 35, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "Model: \"sequential_2\"\n", - "_________________________________________________________________\n", - "Layer (type) Output Shape Param # \n", - "=================================================================\n", - "conv2d_4 (Conv2D) (None, 22, 22, 96) 960 \n", - "_________________________________________________________________\n", - "max_pooling2d_4 (MaxPooling2 (None, 11, 11, 96) 0 \n", - "_________________________________________________________________\n", - "conv2d_5 (Conv2D) (None, 9, 9, 192) 166080 \n", - "_________________________________________________________________\n", - "max_pooling2d_5 (MaxPooling2 (None, 4, 4, 192) 0 \n", - "_________________________________________________________________\n", - "flatten_2 (Flatten) (None, 3072) 0 \n", - "_________________________________________________________________\n", - "dense_6 (Dense) (None, 500) 1536500 \n", - "_________________________________________________________________\n", - "dense_7 (Dense) (None, 500) 250500 \n", - "_________________________________________________________________\n", - "dense_8 (Dense) (None, 43) 21543 \n", - "=================================================================\n", - "Total params: 1,975,583\n", - "Trainable params: 1,975,583\n", - "Non-trainable params: 0\n", - "_________________________________________________________________\n" + "Images of the dataset have this folowing shape : (48, 48, 1)\n" ] } ], "source": [ - "# ---- The model I want to test..\n", - "#\n", - "model = get_model_v1()\n", - "model.summary()\n", - "model.compile(optimizer='adam',\n", - " loss='sparse_categorical_crossentropy',\n", - " metrics=['accuracy'])\n" + "(n,lx,ly,lz) = x_train.shape\n", + "print(\"Images of the dataset have this folowing shape : \",(lx,ly,lz))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "## 4/ Data augmentation" + "**Get and compile a model, with the data shape :**" ] }, { "cell_type": "code", - "execution_count": 34, + "execution_count": 38, "metadata": {}, "outputs": [], "source": [ - "datagen = keras.preprocessing.image.ImageDataGenerator(featurewise_center=False,\n", - " featurewise_std_normalization=False,\n", - " width_shift_range=0.1,\n", - " height_shift_range=0.1,\n", - " zoom_range=0.2,\n", - " shear_range=0.1,\n", - " rotation_range=10.)\n", - "datagen.fit(x_train[:2000])" + "model = get_model_v3(lx,ly,lz)\n", + "\n", + "# model.summary()\n", + "\n", + "model.compile(optimizer='adam',\n", + " loss='sparse_categorical_crossentropy',\n", + " metrics=['accuracy'])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "## 5/ Run model" + "**Train it :** \n", + "Note : La courbe d'apprentissage est visible en temps réel avec Tensorboard : \n", + "`#tensorboard --logdir ./run/logs` " ] }, { "cell_type": "code", - "execution_count": 36, + "execution_count": 40, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "Train on 2000 samples, validate on 200 samples\n", - "Epoch 1/8\n", - "2000/2000 [==============================] - 2s 767us/sample - loss: 0.2712 - accuracy: 0.9195 - val_loss: 1.0751 - val_accuracy: 0.7650\n", - "Epoch 2/8\n", - "2000/2000 [==============================] - 2s 829us/sample - loss: 0.2290 - accuracy: 0.9295 - val_loss: 1.1727 - val_accuracy: 0.7400\n", - "Epoch 3/8\n", - "2000/2000 [==============================] - 2s 906us/sample - loss: 0.1570 - accuracy: 0.9515 - val_loss: 1.0644 - val_accuracy: 0.8350\n", - "Epoch 4/8\n", - "2000/2000 [==============================] - 2s 896us/sample - loss: 0.1282 - accuracy: 0.9640 - val_loss: 1.0879 - val_accuracy: 0.8150\n", - "Epoch 5/8\n", - "2000/2000 [==============================] - 2s 913us/sample - loss: 0.0847 - accuracy: 0.9750 - val_loss: 1.1590 - val_accuracy: 0.8050\n", - "Epoch 6/8\n", - "2000/2000 [==============================] - 2s 920us/sample - loss: 0.0754 - accuracy: 0.9810 - val_loss: 1.2716 - val_accuracy: 0.8100\n", - "Epoch 7/8\n", - "2000/2000 [==============================] - 2s 930us/sample - loss: 0.0910 - accuracy: 0.9750 - val_loss: 1.1533 - val_accuracy: 0.8350\n", - "Epoch 8/8\n", - "2000/2000 [==============================] - 2s 973us/sample - loss: 0.0671 - accuracy: 0.9825 - val_loss: 1.1136 - val_accuracy: 0.8350\n", - "CPU times: user 1min 21s, sys: 7.22 s, total: 1min 29s\n", - "Wall time: 14.3 s\n" + "Train for 613 steps, validate on 12630 samples\n", + "Epoch 1/30\n", + "613/613 [==============================] - 14s 23ms/step - loss: 2.6682 - accuracy: 0.2347 - val_loss: 1.0431 - val_accuracy: 0.6518\n", + "Epoch 2/30\n", + "613/613 [==============================] - 12s 20ms/step - loss: 1.1481 - accuracy: 0.6272 - val_loss: 0.3927 - val_accuracy: 0.8698\n", + "Epoch 3/30\n", + "613/613 [==============================] - 13s 22ms/step - loss: 0.7645 - accuracy: 0.7480 - val_loss: 0.2486 - val_accuracy: 0.9294\n", + "Epoch 4/30\n", + "613/613 [==============================] - 13s 21ms/step - loss: 0.6177 - accuracy: 0.7999 - val_loss: 0.1862 - val_accuracy: 0.9454\n", + "Epoch 5/30\n", + "613/613 [==============================] - 12s 20ms/step - loss: 0.5352 - accuracy: 0.8278 - val_loss: 0.1689 - val_accuracy: 0.9519\n", + "Epoch 6/30\n", + "613/613 [==============================] - 14s 23ms/step - loss: 0.4701 - accuracy: 0.8480 - val_loss: 0.1500 - val_accuracy: 0.9527\n", + "Epoch 7/30\n", + "613/613 [==============================] - 13s 21ms/step - loss: 0.4300 - accuracy: 0.8629 - val_loss: 0.1231 - val_accuracy: 0.9670\n", + "Epoch 8/30\n", + "613/613 [==============================] - 13s 21ms/step - loss: 0.4069 - accuracy: 0.8700 - val_loss: 0.1172 - val_accuracy: 0.9670\n", + "Epoch 9/30\n", + "613/613 [==============================] - 12s 20ms/step - loss: 0.3778 - accuracy: 0.8783 - val_loss: 0.1282 - val_accuracy: 0.9640\n", + "Epoch 10/30\n", + "613/613 [==============================] - 12s 20ms/step - loss: 0.3633 - accuracy: 0.8856 - val_loss: 0.1058 - val_accuracy: 0.9709\n", + "Epoch 11/30\n", + "613/613 [==============================] - 14s 23ms/step - loss: 0.3473 - accuracy: 0.8882 - val_loss: 0.1081 - val_accuracy: 0.9687\n", + "Epoch 12/30\n", + "613/613 [==============================] - 13s 22ms/step - loss: 0.3371 - accuracy: 0.8936 - val_loss: 0.1014 - val_accuracy: 0.9710\n", + "Epoch 13/30\n", + "613/613 [==============================] - 13s 22ms/step - loss: 0.3326 - accuracy: 0.8968 - val_loss: 0.0883 - val_accuracy: 0.9753\n", + "Epoch 14/30\n", + "613/613 [==============================] - 12s 20ms/step - loss: 0.3214 - accuracy: 0.9001 - val_loss: 0.0966 - val_accuracy: 0.9714\n", + "Epoch 15/30\n", + "613/613 [==============================] - 12s 20ms/step - loss: 0.3164 - accuracy: 0.9026 - val_loss: 0.0987 - val_accuracy: 0.9703\n", + "Epoch 16/30\n", + "613/613 [==============================] - 12s 20ms/step - loss: 0.3052 - accuracy: 0.9056 - val_loss: 0.1026 - val_accuracy: 0.9689\n", + "Epoch 17/30\n", + "613/613 [==============================] - 12s 20ms/step - loss: 0.3035 - accuracy: 0.9072 - val_loss: 0.0888 - val_accuracy: 0.9751\n", + "Epoch 18/30\n", + "613/613 [==============================] - 13s 21ms/step - loss: 0.2957 - accuracy: 0.9090 - val_loss: 0.0850 - val_accuracy: 0.9764\n", + "Epoch 19/30\n", + "613/613 [==============================] - 12s 20ms/step - loss: 0.2931 - accuracy: 0.9098 - val_loss: 0.0853 - val_accuracy: 0.9747\n", + "Epoch 20/30\n", + "613/613 [==============================] - 13s 21ms/step - loss: 0.2868 - accuracy: 0.9118 - val_loss: 0.0862 - val_accuracy: 0.9760\n", + "Epoch 21/30\n", + "613/613 [==============================] - 13s 21ms/step - loss: 0.2874 - accuracy: 0.9126 - val_loss: 0.0872 - val_accuracy: 0.9754\n", + "Epoch 22/30\n", + "613/613 [==============================] - 13s 20ms/step - loss: 0.2823 - accuracy: 0.9129 - val_loss: 0.0858 - val_accuracy: 0.9762\n", + "Epoch 23/30\n", + "613/613 [==============================] - 13s 21ms/step - loss: 0.2825 - accuracy: 0.9137 - val_loss: 0.0796 - val_accuracy: 0.9770\n", + "Epoch 24/30\n", + "613/613 [==============================] - 13s 21ms/step - loss: 0.2802 - accuracy: 0.9144 - val_loss: 0.0805 - val_accuracy: 0.9752\n", + "Epoch 25/30\n", + "613/613 [==============================] - 12s 20ms/step - loss: 0.2767 - accuracy: 0.9172 - val_loss: 0.0764 - val_accuracy: 0.9762\n", + "Epoch 26/30\n", + "613/613 [==============================] - 12s 19ms/step - loss: 0.2770 - accuracy: 0.9166 - val_loss: 0.0842 - val_accuracy: 0.9760\n", + "Epoch 27/30\n", + "613/613 [==============================] - 12s 20ms/step - loss: 0.2802 - accuracy: 0.9164 - val_loss: 0.0920 - val_accuracy: 0.9724\n", + "Epoch 28/30\n", + "613/613 [==============================] - 12s 20ms/step - loss: 0.2729 - accuracy: 0.9189 - val_loss: 0.0774 - val_accuracy: 0.9783\n", + "Epoch 29/30\n", + "613/613 [==============================] - 12s 20ms/step - loss: 0.2769 - accuracy: 0.9181 - val_loss: 0.0838 - val_accuracy: 0.9759\n", + "Epoch 30/30\n", + "613/613 [==============================] - 12s 20ms/step - loss: 0.2724 - accuracy: 0.9187 - val_loss: 0.0812 - val_accuracy: 0.9768\n", + "CPU times: user 6min 33s, sys: 11.1 s, total: 6min 45s\n", + "Wall time: 6min 30s\n" ] } ], "source": [ "%%time\n", "\n", + "batch_size = 64\n", + "epochs = 30\n", + "\n", "# ---- Shuffle train data\n", - "# x_train,y_train=ooo.shuffle_np_dataset(x_train,y_train)\n", + "#x_train,y_train=ooo.shuffle_np_dataset(x_train,y_train)\n", "\n", "# ---- Train\n", - "# Note: To be faster in our example, we take only 2000 values\n", - "# but in the real world, we'd take the whole dataset!\n", "#\n", - "history = model.fit( \n", - " x_train[:2000], y_train[:2000], \n", - "# datagen.flow(x_train[:2000], y_train[:2000], batch_size=batch_size),\n", - "# batch_size=batch_size,\n", + "history = model.fit( datagen.flow(x_train, y_train, batch_size=batch_size),\n", " epochs=epochs,\n", " verbose=1,\n", - " validation_data=(x_test[:200], y_test[:200]))\n", + " validation_data=(x_test, y_test),\n", + " callbacks=[tensorboard_callback, bestmodel_callback, savemodel_callback] )\n", "\n", - "# model.save('./run/models/last-model.h5')" + "model.save('./run/models/last-model.h5')" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Evaluate it :**" + ] + }, + { + "cell_type": "code", + "execution_count": 41, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Max validation accuracy is : 0.9783\n" + ] + } + ], + "source": [ + "max_val_accuracy = max(history.history[\"val_accuracy\"])\n", + "print(\"Max validation accuracy is : {:.4f}\".format(max_val_accuracy))" + ] + }, + { + "cell_type": "code", + "execution_count": 42, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Test loss : 0.0812\n", + "Test accuracy : 0.9768\n" + ] + } + ], + "source": [ + "score = model.evaluate(x_test, y_test, verbose=0)\n", + "\n", + "print('Test loss : {:5.4f}'.format(score[0]))\n", + "print('Test accuracy : {:5.4f}'.format(score[1]))" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## 6/ History\n", + "The return of model.fit() returns us the learning history" + ] + }, + { + "cell_type": "code", + "execution_count": 44, + "metadata": { + "jupyter": { + "source_hidden": true + } + }, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "<Figure size 576x432 with 1 Axes>" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "<Figure size 576x432 with 1 Axes>" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "ooo.plot_history(history)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## 8/ Restore and evaluate best model" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### 8.1/ Restore model :" + ] + }, + { + "cell_type": "code", + "execution_count": 45, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Loaded.\n" + ] + } + ], + "source": [ + "loaded_model = tf.keras.models.load_model('./run/models/best-model.h5')\n", + "# best_model.summary()\n", + "print(\"Loaded.\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### 8.2/ Evaluate it :" + ] + }, + { + "cell_type": "code", + "execution_count": 48, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Test loss : 0.0774\n", + "Test accuracy : 0.9783\n" + ] + } + ], + "source": [ + "score = loaded_model.evaluate(x_test, y_test, verbose=0)\n", + "\n", + "print('Test loss : {:5.4f}'.format(score[0]))\n", + "print('Test accuracy : {:5.4f}'.format(score[1]))" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Plot confusion matrix**" + ] + }, + { + "cell_type": "code", + "execution_count": 49, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "<Figure size 864x576 with 2 Axes>" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "y_pred = model.predict_classes(x_test)\n", + "conf_mat = confusion_matrix(y_test,y_pred, normalize=\"true\", labels=range(43))\n", + "\n", + "ooo.plot_confusion_matrix(conf_mat)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "---\n", + "That's all folks !" ] }, { diff --git a/GTSRB/04.1-Data-augmentation.ipynb b/GTSRB/04.1-Data-augmentation.ipynb new file mode 100644 index 0000000000000000000000000000000000000000..cf5ef5807f421a12b5afa8f12bc777c954ecfb90 --- /dev/null +++ b/GTSRB/04.1-Data-augmentation.ipynb @@ -0,0 +1,353 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "German Traffic Sign Recognition Benchmark (GTSRB)\n", + "=================================================\n", + "---\n", + "Introduction au Deep Learning (IDLE) - S. Aria, E. Maldonado, JL. Parouty - CNRS/SARI/DEVLOG - 2020\n", + "\n", + "## Episode 3 : Tracking, visualizing and save models\n", + "\n", + "Our main steps:\n", + " - Monitoring and understanding our model training\n", + " - Analyze the results \n", + " - Improving our model\n", + " - Add recovery points\n", + "\n", + "\n", + "## 1/ Import and init" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "IDLE 2020 - Practical Work Module\n", + " Version : 0.1.4\n", + " Run time : Monday 13 January 2020, 21:13:18\n", + " Matplotlib style : idle/talk.mplstyle\n", + " TensorFlow version : 2.0.0\n", + " Keras version : 2.2.4-tf\n" + ] + } + ], + "source": [ + "import tensorflow as tf\n", + "from tensorflow import keras\n", + "from tensorflow.keras.callbacks import TensorBoard\n", + "\n", + "import numpy as np\n", + "import h5py\n", + "import matplotlib.pyplot as plt\n", + "import os, time, random\n", + "\n", + "import idle.pwk as ooo\n", + "from importlib import reload\n", + "\n", + "ooo.init()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## 2/ Reload dataset\n", + "Dataset is one of the saved dataset: RGB25, RGB35, L25, L35, etc. \n", + "First of all, we're going to use a smart dataset : **set-24x24-L** \n", + "(with a GPU, it only takes 35'' compared to more than 5' with a CPU !)" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Dataset loaded (228.8 Mo)\n", + "\n", + "CPU times: user 0 ns, sys: 297 ms, total: 297 ms\n", + "Wall time: 294 ms\n" + ] + } + ], + "source": [ + "%%time\n", + "\n", + "dataset ='set-24x24-L'\n", + "\n", + "# ---- Read dataset\n", + "#\n", + "filename='./data/'+dataset+'.h5'\n", + "with h5py.File(filename) as f:\n", + " x_train = f['x_train'][:]\n", + " y_train = f['y_train'][:]\n", + " x_test = f['x_test'][:]\n", + " y_test = f['y_test'][:]\n", + "\n", + "# ---- Dataset shape\n", + "#\n", + "(n,lx,ly,lz) = x_train.shape\n", + "data_shape = (lx, ly, lz)\n", + "\n", + "# ---- done\n", + "print('Dataset loaded ({:.1f} Mo)\\n'.format(os.path.getsize(filename)/(1024*1024)))" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## 3/ Have a look to the dataset\n", + "Note: Data must be reshape for matplotlib" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "x_train : (39209, 24, 24, 1)\n", + "y_train : (39209,)\n", + "x_test : (12630, 24, 24, 1)\n", + "y_test : (12630,)\n" + ] + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "<Figure size 864x291.6 with 36 Axes>" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "print(\"x_train : \", x_train.shape)\n", + "print(\"y_train : \", y_train.shape)\n", + "print(\"x_test : \", x_test.shape)\n", + "print(\"y_test : \", y_test.shape)\n", + "\n", + "ooo.plot_images(x_train, y_train, range(36), columns=12, x_size=1, y_size=1)\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## 4/ Create model\n", + "Some hyperparameters :" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "metadata": {}, + "outputs": [], + "source": [ + "batch_size = 64\n", + "num_classes = 43\n", + "epochs = 8" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "My models :" + ] + }, + { + "cell_type": "code", + "execution_count": 32, + "metadata": {}, + "outputs": [], + "source": [ + "\n", + "def get_model_v1():\n", + " model = keras.models.Sequential()\n", + " model.add( keras.layers.Conv2D(96, (3,3), activation='relu', input_shape=(lx,ly,lz)))\n", + " model.add( keras.layers.MaxPooling2D((2, 2)))\n", + " model.add( keras.layers.Conv2D(192, (3, 3), activation='relu'))\n", + " model.add( keras.layers.MaxPooling2D((2, 2)))\n", + " model.add( keras.layers.Flatten()) \n", + " model.add( keras.layers.Dense(500, activation='relu'))\n", + " model.add( keras.layers.Dense(500, activation='relu'))\n", + " model.add( keras.layers.Dense(43, activation='softmax'))\n", + " return model\n" + ] + }, + { + "cell_type": "code", + "execution_count": 33, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Model: \"sequential_2\"\n", + "_________________________________________________________________\n", + "Layer (type) Output Shape Param # \n", + "=================================================================\n", + "conv2d_4 (Conv2D) (None, 22, 22, 96) 960 \n", + "_________________________________________________________________\n", + "max_pooling2d_4 (MaxPooling2 (None, 11, 11, 96) 0 \n", + "_________________________________________________________________\n", + "conv2d_5 (Conv2D) (None, 9, 9, 192) 166080 \n", + "_________________________________________________________________\n", + "max_pooling2d_5 (MaxPooling2 (None, 4, 4, 192) 0 \n", + "_________________________________________________________________\n", + "flatten_2 (Flatten) (None, 3072) 0 \n", + "_________________________________________________________________\n", + "dense_6 (Dense) (None, 500) 1536500 \n", + "_________________________________________________________________\n", + "dense_7 (Dense) (None, 500) 250500 \n", + "_________________________________________________________________\n", + "dense_8 (Dense) (None, 43) 21543 \n", + "=================================================================\n", + "Total params: 1,975,583\n", + "Trainable params: 1,975,583\n", + "Non-trainable params: 0\n", + "_________________________________________________________________\n" + ] + } + ], + "source": [ + "# ---- The model I want to test..\n", + "#\n", + "model = get_model_v1()\n", + "model.summary()\n", + "model.compile(optimizer='adam',\n", + " loss='sparse_categorical_crossentropy',\n", + " metrics=['accuracy'])\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## 4/ Data augmentation" + ] + }, + { + "cell_type": "code", + "execution_count": 34, + "metadata": {}, + "outputs": [], + "source": [ + "datagen = keras.preprocessing.image.ImageDataGenerator(featurewise_center=False,\n", + " featurewise_std_normalization=False,\n", + " width_shift_range=0.1,\n", + " height_shift_range=0.1,\n", + " zoom_range=0.2,\n", + " shear_range=0.1,\n", + " rotation_range=10.)\n", + "datagen.fit(x_train[:2000])" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## 5/ Run model" + ] + }, + { + "cell_type": "code", + "execution_count": 36, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Train on 2000 samples, validate on 200 samples\n", + "Epoch 1/8\n", + "2000/2000 [==============================] - 2s 767us/sample - loss: 0.2712 - accuracy: 0.9195 - val_loss: 1.0751 - val_accuracy: 0.7650\n", + "Epoch 2/8\n", + "2000/2000 [==============================] - 2s 829us/sample - loss: 0.2290 - accuracy: 0.9295 - val_loss: 1.1727 - val_accuracy: 0.7400\n", + "Epoch 3/8\n", + "2000/2000 [==============================] - 2s 906us/sample - loss: 0.1570 - accuracy: 0.9515 - val_loss: 1.0644 - val_accuracy: 0.8350\n", + "Epoch 4/8\n", + "2000/2000 [==============================] - 2s 896us/sample - loss: 0.1282 - accuracy: 0.9640 - val_loss: 1.0879 - val_accuracy: 0.8150\n", + "Epoch 5/8\n", + "2000/2000 [==============================] - 2s 913us/sample - loss: 0.0847 - accuracy: 0.9750 - val_loss: 1.1590 - val_accuracy: 0.8050\n", + "Epoch 6/8\n", + "2000/2000 [==============================] - 2s 920us/sample - loss: 0.0754 - accuracy: 0.9810 - val_loss: 1.2716 - val_accuracy: 0.8100\n", + "Epoch 7/8\n", + "2000/2000 [==============================] - 2s 930us/sample - loss: 0.0910 - accuracy: 0.9750 - val_loss: 1.1533 - val_accuracy: 0.8350\n", + "Epoch 8/8\n", + "2000/2000 [==============================] - 2s 973us/sample - loss: 0.0671 - accuracy: 0.9825 - val_loss: 1.1136 - val_accuracy: 0.8350\n", + "CPU times: user 1min 21s, sys: 7.22 s, total: 1min 29s\n", + "Wall time: 14.3 s\n" + ] + } + ], + "source": [ + "%%time\n", + "\n", + "# ---- Shuffle train data\n", + "# x_train,y_train=ooo.shuffle_np_dataset(x_train,y_train)\n", + "\n", + "# ---- Train\n", + "# Note: To be faster in our example, we take only 2000 values\n", + "# but in the real world, we'd take the whole dataset!\n", + "#\n", + "history = model.fit( \n", + " x_train[:2000], y_train[:2000], \n", + "# datagen.flow(x_train[:2000], y_train[:2000], batch_size=batch_size),\n", + "# batch_size=batch_size,\n", + " epochs=epochs,\n", + " verbose=1,\n", + " validation_data=(x_test[:200], y_test[:200]))\n", + "\n", + "# model.save('./run/models/last-model.h5')" + ] + }, + { + "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": "3.7.5" + } + }, + "nbformat": 4, + "nbformat_minor": 4 +} diff --git a/GTSRB/99 Run-Tensorboard.ipynb b/GTSRB/99 Run-Tensorboard.ipynb index 162e91a647f34a90b34fe051a51ecf7e21b8c657..0b798b383e2e383eeef37a34e94a6db5536a2082 100644 --- a/GTSRB/99 Run-Tensorboard.ipynb +++ b/GTSRB/99 Run-Tensorboard.ipynb @@ -21,14 +21,14 @@ }, { "cell_type": "code", - "execution_count": 220, + "execution_count": 1, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "Tensorbord started with pid 84593\n" + "Tensorbord started with pid 80451\n" ] } ], @@ -57,9 +57,50 @@ }, { "cell_type": "code", - "execution_count": 79, + "execution_count": 19, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "W0113 22:02:28.498013 140212267140864 plugin_event_accumulator.py:294] Found more than one graph event per run, or there was a metagraph containing a graph_def, as well as one or more graph events. Overwriting the graph with the newest event.\n", + "TensorBoard 2.0.0 at http://0.0.0.0:21277/ (Press CTRL+C to quit)\n", + "Traceback (most recent call last):\n", + " File \"/home/paroutyj/.conda/envs/deeplearning2/bin/tensorboard\", line 10, in <module>\n", + " sys.exit(run_main())\n", + " File \"/home/paroutyj/.conda/envs/deeplearning2/lib/python3.7/site-packages/tensorboard/main.py\", line 64, in run_main\n", + " app.run(tensorboard.main, flags_parser=tensorboard.configure)\n", + " File \"/home/paroutyj/.conda/envs/deeplearning2/lib/python3.7/site-packages/absl/app.py\", line 299, in run\n", + " _run_main(main, args)\n", + " File \"/home/paroutyj/.conda/envs/deeplearning2/lib/python3.7/site-packages/absl/app.py\", line 250, in _run_main\n", + " sys.exit(main(argv))\n", + " File \"/home/paroutyj/.conda/envs/deeplearning2/lib/python3.7/site-packages/tensorboard/program.py\", line 222, in main\n", + " self._register_info(server)\n", + " File \"/home/paroutyj/.conda/envs/deeplearning2/lib/python3.7/site-packages/tensorboard/program.py\", line 268, in _register_info\n", + " manager.write_info_file(info)\n", + " File \"/home/paroutyj/.conda/envs/deeplearning2/lib/python3.7/site-packages/tensorboard/manager.py\", line 268, in write_info_file\n", + " with open(_get_info_file_path(), \"w\") as outfile:\n", + "PermissionError: [Errno 13] Permission denied: '/tmp/.tensorboard-info/pid-94212.info'\n" + ] + }, + { + "ename": "CalledProcessError", + "evalue": "Command 'b'\\n# ---- Port number, \\nPORT_JPY=\"$(id -u)\"\\nPORT_TSB=\"$(( $PORT_J + 10000 ))\"\\nHOST_G=\"$(hostname)\"\\nSSH_CMD=\"/usr/bin/ssh -NL 8888:$HOST_G:$PORT_J -L 6006:$HOST_G:$PORT_T dahu.ciment\"\\n\\n# tensorboard --port 21277 --host 0.0.0.0 --logdir ./run/logs &>/dev/null &\\ntensorboard --port 21277 --host 0.0.0.0 --logdir ./run/logs 2>&1\\n'' returned non-zero exit status 1.", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mCalledProcessError\u001b[0m Traceback (most recent call last)", + "\u001b[0;32m<ipython-input-19-d5c18c938787>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mget_ipython\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrun_cell_magic\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'bash'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m''\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'\\n# ---- Port number, \\nPORT_JPY=\"$(id -u)\"\\nPORT_TSB=\"$(( $PORT_J + 10000 ))\"\\nHOST_G=\"$(hostname)\"\\nSSH_CMD=\"/usr/bin/ssh -NL 8888:$HOST_G:$PORT_J -L 6006:$HOST_G:$PORT_T dahu.ciment\"\\n\\n# tensorboard --port 21277 --host 0.0.0.0 --logdir ./run/logs &>/dev/null &\\ntensorboard --port 21277 --host 0.0.0.0 --logdir ./run/logs 2>&1\\n'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m", + "\u001b[0;32m~/.conda/envs/deeplearning2/lib/python3.7/site-packages/IPython/core/interactiveshell.py\u001b[0m in \u001b[0;36mrun_cell_magic\u001b[0;34m(self, magic_name, line, cell)\u001b[0m\n\u001b[1;32m 2350\u001b[0m \u001b[0;32mwith\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbuiltin_trap\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2351\u001b[0m \u001b[0margs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mmagic_arg_s\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcell\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2352\u001b[0;31m \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 2353\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mresult\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2354\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/.conda/envs/deeplearning2/lib/python3.7/site-packages/IPython/core/magics/script.py\u001b[0m in \u001b[0;36mnamed_script_magic\u001b[0;34m(line, cell)\u001b[0m\n\u001b[1;32m 140\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 141\u001b[0m \u001b[0mline\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mscript\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 142\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshebang\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mline\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcell\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 143\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 144\u001b[0m \u001b[0;31m# write a basic docstring:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m</home/paroutyj/.conda/envs/deeplearning2/lib/python3.7/site-packages/decorator.py:decorator-gen-110>\u001b[0m in \u001b[0;36mshebang\u001b[0;34m(self, line, cell)\u001b[0m\n", + "\u001b[0;32m~/.conda/envs/deeplearning2/lib/python3.7/site-packages/IPython/core/magic.py\u001b[0m in \u001b[0;36m<lambda>\u001b[0;34m(f, *a, **k)\u001b[0m\n\u001b[1;32m 185\u001b[0m \u001b[0;31m# but it's overkill for just that one bit of state.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 186\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mmagic_deco\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0marg\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 187\u001b[0;31m \u001b[0mcall\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mlambda\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0ma\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mk\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0ma\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mk\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 188\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 189\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mcallable\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0marg\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/.conda/envs/deeplearning2/lib/python3.7/site-packages/IPython/core/magics/script.py\u001b[0m in \u001b[0;36mshebang\u001b[0;34m(self, line, cell)\u001b[0m\n\u001b[1;32m 243\u001b[0m \u001b[0msys\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstderr\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mflush\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 244\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mraise_error\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mreturncode\u001b[0m\u001b[0;34m!=\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 245\u001b[0;31m \u001b[0;32mraise\u001b[0m \u001b[0mCalledProcessError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mreturncode\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcell\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0moutput\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mout\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstderr\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0merr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 246\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 247\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0m_run_script\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mp\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcell\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mto_close\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;31mCalledProcessError\u001b[0m: Command 'b'\\n# ---- Port number, \\nPORT_JPY=\"$(id -u)\"\\nPORT_TSB=\"$(( $PORT_J + 10000 ))\"\\nHOST_G=\"$(hostname)\"\\nSSH_CMD=\"/usr/bin/ssh -NL 8888:$HOST_G:$PORT_J -L 6006:$HOST_G:$PORT_T dahu.ciment\"\\n\\n# tensorboard --port 21277 --host 0.0.0.0 --logdir ./run/logs &>/dev/null &\\ntensorboard --port 21277 --host 0.0.0.0 --logdir ./run/logs 2>&1\\n'' returned non-zero exit status 1." + ] + } + ], "source": [ "%%bash\n", "\n", @@ -69,7 +110,34 @@ "HOST_G=\"$(hostname)\"\n", "SSH_CMD=\"/usr/bin/ssh -NL 8888:$HOST_G:$PORT_J -L 6006:$HOST_G:$PORT_T dahu.ciment\"\n", "\n", - "tensorboard --port 21277 --host 0.0.0.0 --logdir ./run/logs &>/dev/null &" + "# tensorboard --port 21277 --host 0.0.0.0 --logdir ./run/logs &>/dev/null &\n", + "tensorboard --port 21277 --host 0.0.0.0 --logdir ./run/logs 2>&1" + ] + }, + { + "cell_type": "code", + "execution_count": 21, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "/bin/sh: 1: cannot create /tmp/.tensorboard-info/foobar-655465.txt: Permission denied\n", + "drwxr-xr-x 2 juanbaldonado l-inria 60 Jan 13 21:00 /tmp/.tensorboard-info\n", + "-rw-r--r-- 1 juanbaldonado l-inria 346 Jan 13 21:00 /tmp/.tensorboard-info/pid-74153.info\n" + ] + } + ], + "source": [ + "!echo \"This a test\">/tmp/.tensorboard-info/foobar-655465.txt\n", + "#!cat /tmp/foobar-655465.txt\n", + "#!rm /tmp/foobar-655465.txt\n", + "!ls -ld /tmp/.tensorboard-info\n", + "!ls -ld /tmp/.tensorboard-info/*\n", + "\n", + "\n", + "See : https://github.com/tensorflow/tensorboard/issues/2010" ] }, { @@ -81,14 +149,14 @@ }, { "cell_type": "code", - "execution_count": 223, + "execution_count": 6, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "84593\n" + "\n" ] } ],