{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"name":"5-3 사전 훈련된 컨브넷.ipynb","provenance":[],"collapsed_sections":[],"authorship_tag":"ABX9TyNp5S7c/N8cjfEMZfPnOFgE"},"kernelspec":{"name":"python3","display_name":"Python 3"},"accelerator":"GPU"},"cells":[{"cell_type":"code","metadata":{"id":"X06r6Jh1Tt9p","colab_type":"code","outputId":"063f51ee-2d9a-42f3-def0-e3f12be68179","executionInfo":{"status":"ok","timestamp":1580353879782,"user_tz":-540,"elapsed":24034,"user":{"displayName":"Yoon Jack","photoUrl":"","userId":"04923927567667044980"}},"colab":{"base_uri":"https://localhost:8080/","height":127}},"source":["from google.colab import drive\n","drive.mount('/content/drive')"],"execution_count":1,"outputs":[{"output_type":"stream","text":["Go to this URL in a browser: https://accounts.google.com/o/oauth2/auth?client_id=947318989803-6bn6qk8qdgf4n4g3pfee6491hc0brc4i.apps.googleusercontent.com&redirect_uri=urn%3aietf%3awg%3aoauth%3a2.0%3aoob&response_type=code&scope=email%20https%3a%2f%2fwww.googleapis.com%2fauth%2fdocs.test%20https%3a%2f%2fwww.googleapis.com%2fauth%2fdrive%20https%3a%2f%2fwww.googleapis.com%2fauth%2fdrive.photos.readonly%20https%3a%2f%2fwww.googleapis.com%2fauth%2fpeopleapi.readonly\n","\n","Enter your authorization code:\n","··········\n","Mounted at /content/drive\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"r0YPI3DOizww","colab_type":"code","outputId":"9738d9c3-8a2b-4687-9e14-46c3203a05c1","executionInfo":{"status":"ok","timestamp":1580353893276,"user_tz":-540,"elapsed":3339,"user":{"displayName":"Yoon Jack","photoUrl":"","userId":"04923927567667044980"}},"colab":{"base_uri":"https://localhost:8080/","height":71}},"source":["!ls -ltr"],"execution_count":2,"outputs":[{"output_type":"stream","text":["total 8\n","drwxr-xr-x 1 root root 4096 Jan 13 16:38 sample_data\n","drwx------ 4 root root 4096 Jan 30 03:11 drive\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"qKOR-SZfi_rD","colab_type":"code","outputId":"6bf15312-78fd-4bc0-d93a-b704f7f98f71","executionInfo":{"status":"ok","timestamp":1580353894071,"user_tz":-540,"elapsed":723,"user":{"displayName":"Yoon Jack","photoUrl":"","userId":"04923927567667044980"}},"colab":{"base_uri":"https://localhost:8080/","height":35}},"source":["cd drive"],"execution_count":3,"outputs":[{"output_type":"stream","text":["/content/drive\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"LbjotCR9jDSj","colab_type":"code","outputId":"30613464-faa7-40c0-fa8c-b7cb9cbea8d6","executionInfo":{"status":"ok","timestamp":1580353894947,"user_tz":-540,"elapsed":393,"user":{"displayName":"Yoon Jack","photoUrl":"","userId":"04923927567667044980"}},"colab":{"base_uri":"https://localhost:8080/","height":35}},"source":["cd 'My Drive'/'Colab Notebooks'/Keras_creator"],"execution_count":4,"outputs":[{"output_type":"stream","text":["/content/drive/My Drive/Colab Notebooks/Keras_creator\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"H0gbluS6jEkS","colab_type":"code","outputId":"89add21d-02a1-4a93-b722-9948db64c8d4","executionInfo":{"status":"ok","timestamp":1580353897975,"user_tz":-540,"elapsed":2321,"user":{"displayName":"Yoon Jack","photoUrl":"","userId":"04923927567667044980"}},"colab":{"base_uri":"https://localhost:8080/","height":118}},"source":["import keras\n","keras.__version__"],"execution_count":5,"outputs":[{"output_type":"stream","text":["Using TensorFlow backend.\n"],"name":"stderr"},{"output_type":"display_data","data":{"text/html":["<p style=\"color: red;\">\n","The default version of TensorFlow in Colab will soon switch to TensorFlow 2.x.<br>\n","We recommend you <a href=\"https://www.tensorflow.org/guide/migrate\" target=\"_blank\">upgrade</a> now \n","or ensure your notebook will continue to use TensorFlow 1.x via the <code>%tensorflow_version 1.x</code> magic:\n","<a href=\"https://colab.research.google.com/notebooks/tensorflow_version.ipynb\" target=\"_blank\">more info</a>.</p>\n"],"text/plain":["<IPython.core.display.HTML object>"]},"metadata":{"tags":[]}},{"output_type":"execute_result","data":{"text/plain":["'2.2.5'"]},"metadata":{"tags":[]},"execution_count":5}]},{"cell_type":"code","metadata":{"id":"r7A0hV8FjIbR","colab_type":"code","outputId":"3241e4b1-8297-416c-8fd7-e57f02e179da","executionInfo":{"status":"ok","timestamp":1580353909187,"user_tz":-540,"elapsed":10521,"user":{"displayName":"Yoon Jack","photoUrl":"","userId":"04923927567667044980"}},"colab":{"base_uri":"https://localhost:8080/","height":433}},"source":["from keras.applications import VGG16\n","\n","conv_base = VGG16(weights='imagenet',\n","                  include_top=False,\n","                  input_shape=(150, 150, 3))"],"execution_count":6,"outputs":[{"output_type":"stream","text":["WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:66: The name tf.get_default_graph is deprecated. Please use tf.compat.v1.get_default_graph instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:541: The name tf.placeholder is deprecated. Please use tf.compat.v1.placeholder instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:4432: The name tf.random_uniform is deprecated. Please use tf.random.uniform instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:4267: The name tf.nn.max_pool is deprecated. Please use tf.nn.max_pool2d instead.\n","\n","Downloading data from https://github.com/fchollet/deep-learning-models/releases/download/v0.1/vgg16_weights_tf_dim_ordering_tf_kernels_notop.h5\n","58892288/58889256 [==============================] - 1s 0us/step\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:190: The name tf.get_default_session is deprecated. Please use tf.compat.v1.get_default_session instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:197: The name tf.ConfigProto is deprecated. Please use tf.compat.v1.ConfigProto instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:203: The name tf.Session is deprecated. Please use tf.compat.v1.Session instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:207: The name tf.global_variables is deprecated. Please use tf.compat.v1.global_variables instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:216: The name tf.is_variable_initialized is deprecated. Please use tf.compat.v1.is_variable_initialized instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:223: The name tf.variables_initializer is deprecated. Please use tf.compat.v1.variables_initializer instead.\n","\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"vjiDL-bTjKFI","colab_type":"code","outputId":"13105971-b015-43b9-8b67-1e007cbd11ea","executionInfo":{"status":"ok","timestamp":1580353913165,"user_tz":-540,"elapsed":723,"user":{"displayName":"Yoon Jack","photoUrl":"","userId":"04923927567667044980"}},"colab":{"base_uri":"https://localhost:8080/","height":845}},"source":["conv_base.summary()"],"execution_count":7,"outputs":[{"output_type":"stream","text":["Model: \"vgg16\"\n","_________________________________________________________________\n","Layer (type)                 Output Shape              Param #   \n","=================================================================\n","input_1 (InputLayer)         (None, 150, 150, 3)       0         \n","_________________________________________________________________\n","block1_conv1 (Conv2D)        (None, 150, 150, 64)      1792      \n","_________________________________________________________________\n","block1_conv2 (Conv2D)        (None, 150, 150, 64)      36928     \n","_________________________________________________________________\n","block1_pool (MaxPooling2D)   (None, 75, 75, 64)        0         \n","_________________________________________________________________\n","block2_conv1 (Conv2D)        (None, 75, 75, 128)       73856     \n","_________________________________________________________________\n","block2_conv2 (Conv2D)        (None, 75, 75, 128)       147584    \n","_________________________________________________________________\n","block2_pool (MaxPooling2D)   (None, 37, 37, 128)       0         \n","_________________________________________________________________\n","block3_conv1 (Conv2D)        (None, 37, 37, 256)       295168    \n","_________________________________________________________________\n","block3_conv2 (Conv2D)        (None, 37, 37, 256)       590080    \n","_________________________________________________________________\n","block3_conv3 (Conv2D)        (None, 37, 37, 256)       590080    \n","_________________________________________________________________\n","block3_pool (MaxPooling2D)   (None, 18, 18, 256)       0         \n","_________________________________________________________________\n","block4_conv1 (Conv2D)        (None, 18, 18, 512)       1180160   \n","_________________________________________________________________\n","block4_conv2 (Conv2D)        (None, 18, 18, 512)       2359808   \n","_________________________________________________________________\n","block4_conv3 (Conv2D)        (None, 18, 18, 512)       2359808   \n","_________________________________________________________________\n","block4_pool (MaxPooling2D)   (None, 9, 9, 512)         0         \n","_________________________________________________________________\n","block5_conv1 (Conv2D)        (None, 9, 9, 512)         2359808   \n","_________________________________________________________________\n","block5_conv2 (Conv2D)        (None, 9, 9, 512)         2359808   \n","_________________________________________________________________\n","block5_conv3 (Conv2D)        (None, 9, 9, 512)         2359808   \n","_________________________________________________________________\n","block5_pool (MaxPooling2D)   (None, 4, 4, 512)         0         \n","=================================================================\n","Total params: 14,714,688\n","Trainable params: 14,714,688\n","Non-trainable params: 0\n","_________________________________________________________________\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"c8RiTWLHjLxO","colab_type":"code","colab":{}},"source":["import os\n","import numpy as np\n","from keras.preprocessing.image import ImageDataGenerator\n","\n","base_dir = './datasets/cats_and_dogs_small'\n","\n","train_dir = os.path.join(base_dir, 'train')\n","validation_dir = os.path.join(base_dir, 'validation')\n","test_dir = os.path.join(base_dir, 'test')\n","\n","datagen = ImageDataGenerator(rescale=1./255)\n","batch_size = 20\n","\n","def extract_features(directory, sample_count):\n","    features = np.zeros(shape=(sample_count, 4, 4, 512))\n","    labels = np.zeros(shape=(sample_count))\n","    generator = datagen.flow_from_directory(\n","        directory,\n","        target_size=(150, 150),\n","        batch_size=batch_size,\n","        class_mode='binary')\n","    i = 0\n","    for inputs_batch, labels_batch in generator:\n","        features_batch = conv_base.predict(inputs_batch)\n","        features[i * batch_size : (i + 1) * batch_size] = features_batch\n","        labels[i * batch_size : (i + 1) * batch_size] = labels_batch\n","        i += 1\n","        if i * batch_size >= sample_count:\n","            # 제너레이터는 루프 안에서 무한하게 데이터를 만들어내므로 모든 이미지를 한 번씩 처리하고 나면 중지합니다\n","            break\n","    return features, labels"],"execution_count":0,"outputs":[]},{"cell_type":"code","metadata":{"id":"n0RVLSGGjPbA","colab_type":"code","outputId":"323e4a45-4080-475c-fb61-c4211891d7c3","executionInfo":{"status":"ok","timestamp":1580354773931,"user_tz":-540,"elapsed":857786,"user":{"displayName":"Yoon Jack","photoUrl":"","userId":"04923927567667044980"}},"colab":{"base_uri":"https://localhost:8080/","height":71}},"source":["train_features, train_labels = extract_features(train_dir, 2000)\n","validation_features, validation_labels = extract_features(validation_dir, 1000)\n","test_features, test_labels = extract_features(test_dir, 1000)"],"execution_count":9,"outputs":[{"output_type":"stream","text":["Found 2000 images belonging to 2 classes.\n","Found 1000 images belonging to 2 classes.\n","Found 1000 images belonging to 2 classes.\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"EMBVR6KNjRWQ","colab_type":"code","colab":{}},"source":["train_features = np.reshape(train_features, (2000, 4 * 4 * 512))\n","validation_features = np.reshape(validation_features, (1000, 4 * 4 * 512))\n","test_features = np.reshape(test_features, (1000, 4 * 4 * 512))"],"execution_count":0,"outputs":[]},{"cell_type":"code","metadata":{"id":"8jGVX6vaqaXs","colab_type":"code","outputId":"1c7e431a-dbab-4d61-8bbd-57c6520cc95b","executionInfo":{"status":"ok","timestamp":1580354940397,"user_tz":-540,"elapsed":18736,"user":{"displayName":"Yoon Jack","photoUrl":"","userId":"04923927567667044980"}},"colab":{"base_uri":"https://localhost:8080/","height":1000}},"source":["from keras import models\n","from keras import layers\n","from keras import optimizers\n","\n","model = models.Sequential()\n","model.add(layers.Dense(256, activation='relu', input_dim=4 * 4 * 512))\n","model.add(layers.Dropout(0.5))\n","model.add(layers.Dense(1, activation='sigmoid'))\n","\n","model.compile(optimizer=optimizers.RMSprop(lr=2e-5),\n","              loss='binary_crossentropy',\n","              metrics=['acc'])\n","\n","history = model.fit(train_features, train_labels,\n","                    epochs=30,\n","                    batch_size=20,\n","                    validation_data=(validation_features, validation_labels))"],"execution_count":11,"outputs":[{"output_type":"stream","text":["WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:148: The name tf.placeholder_with_default is deprecated. Please use tf.compat.v1.placeholder_with_default instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:3733: calling dropout (from tensorflow.python.ops.nn_ops) with keep_prob is deprecated and will be removed in a future version.\n","Instructions for updating:\n","Please use `rate` instead of `keep_prob`. Rate should be set to `rate = 1 - keep_prob`.\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/optimizers.py:793: The name tf.train.Optimizer is deprecated. Please use tf.compat.v1.train.Optimizer instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:3657: The name tf.log is deprecated. Please use tf.math.log instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow_core/python/ops/nn_impl.py:183: where (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version.\n","Instructions for updating:\n","Use tf.where in 2.0, which has the same broadcast rule as np.where\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:1033: The name tf.assign_add is deprecated. Please use tf.compat.v1.assign_add instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:1020: The name tf.assign is deprecated. Please use tf.compat.v1.assign instead.\n","\n","Train on 2000 samples, validate on 1000 samples\n","Epoch 1/30\n","2000/2000 [==============================] - 1s 460us/step - loss: 0.6049 - acc: 0.6665 - val_loss: 0.4293 - val_acc: 0.8450\n","Epoch 2/30\n","2000/2000 [==============================] - 1s 288us/step - loss: 0.4274 - acc: 0.8065 - val_loss: 0.3503 - val_acc: 0.8730\n","Epoch 3/30\n","2000/2000 [==============================] - 1s 288us/step - loss: 0.3469 - acc: 0.8560 - val_loss: 0.3208 - val_acc: 0.8720\n","Epoch 4/30\n","2000/2000 [==============================] - 1s 292us/step - loss: 0.3086 - acc: 0.8775 - val_loss: 0.2912 - val_acc: 0.8910\n","Epoch 5/30\n","2000/2000 [==============================] - 1s 287us/step - loss: 0.2909 - acc: 0.8770 - val_loss: 0.2769 - val_acc: 0.8980\n","Epoch 6/30\n","2000/2000 [==============================] - 1s 289us/step - loss: 0.2587 - acc: 0.8970 - val_loss: 0.2651 - val_acc: 0.9030\n","Epoch 7/30\n","2000/2000 [==============================] - 1s 289us/step - loss: 0.2395 - acc: 0.9055 - val_loss: 0.2601 - val_acc: 0.8960\n","Epoch 8/30\n","2000/2000 [==============================] - 1s 295us/step - loss: 0.2249 - acc: 0.9120 - val_loss: 0.2513 - val_acc: 0.9030\n","Epoch 9/30\n","2000/2000 [==============================] - 1s 280us/step - loss: 0.2140 - acc: 0.9210 - val_loss: 0.2572 - val_acc: 0.8930\n","Epoch 10/30\n","2000/2000 [==============================] - 1s 283us/step - loss: 0.1973 - acc: 0.9280 - val_loss: 0.2409 - val_acc: 0.9040\n","Epoch 11/30\n","2000/2000 [==============================] - 1s 290us/step - loss: 0.1980 - acc: 0.9265 - val_loss: 0.2390 - val_acc: 0.9080\n","Epoch 12/30\n","2000/2000 [==============================] - 1s 298us/step - loss: 0.1872 - acc: 0.9310 - val_loss: 0.2385 - val_acc: 0.9060\n","Epoch 13/30\n","2000/2000 [==============================] - 1s 312us/step - loss: 0.1732 - acc: 0.9385 - val_loss: 0.2374 - val_acc: 0.9070\n","Epoch 14/30\n","2000/2000 [==============================] - 1s 291us/step - loss: 0.1637 - acc: 0.9435 - val_loss: 0.2389 - val_acc: 0.8970\n","Epoch 15/30\n","2000/2000 [==============================] - 1s 296us/step - loss: 0.1583 - acc: 0.9490 - val_loss: 0.2367 - val_acc: 0.9040\n","Epoch 16/30\n","2000/2000 [==============================] - 1s 307us/step - loss: 0.1505 - acc: 0.9485 - val_loss: 0.2342 - val_acc: 0.9110\n","Epoch 17/30\n","2000/2000 [==============================] - 1s 296us/step - loss: 0.1474 - acc: 0.9485 - val_loss: 0.2316 - val_acc: 0.9070\n","Epoch 18/30\n","2000/2000 [==============================] - 1s 287us/step - loss: 0.1402 - acc: 0.9490 - val_loss: 0.2371 - val_acc: 0.9010\n","Epoch 19/30\n","2000/2000 [==============================] - 1s 283us/step - loss: 0.1369 - acc: 0.9545 - val_loss: 0.2459 - val_acc: 0.8930\n","Epoch 20/30\n","2000/2000 [==============================] - 1s 293us/step - loss: 0.1281 - acc: 0.9555 - val_loss: 0.2315 - val_acc: 0.9090\n","Epoch 21/30\n","2000/2000 [==============================] - 1s 290us/step - loss: 0.1224 - acc: 0.9605 - val_loss: 0.2307 - val_acc: 0.9050\n","Epoch 22/30\n","2000/2000 [==============================] - 1s 288us/step - loss: 0.1195 - acc: 0.9620 - val_loss: 0.2308 - val_acc: 0.9080\n","Epoch 23/30\n","2000/2000 [==============================] - 1s 282us/step - loss: 0.1099 - acc: 0.9665 - val_loss: 0.2350 - val_acc: 0.9080\n","Epoch 24/30\n","2000/2000 [==============================] - 1s 308us/step - loss: 0.1096 - acc: 0.9645 - val_loss: 0.2473 - val_acc: 0.8920\n","Epoch 25/30\n","2000/2000 [==============================] - 1s 301us/step - loss: 0.1052 - acc: 0.9655 - val_loss: 0.2320 - val_acc: 0.9060\n","Epoch 26/30\n","2000/2000 [==============================] - 1s 286us/step - loss: 0.0972 - acc: 0.9710 - val_loss: 0.2348 - val_acc: 0.9060\n","Epoch 27/30\n","2000/2000 [==============================] - 1s 296us/step - loss: 0.0974 - acc: 0.9700 - val_loss: 0.2370 - val_acc: 0.9070\n","Epoch 28/30\n","2000/2000 [==============================] - 1s 281us/step - loss: 0.0932 - acc: 0.9735 - val_loss: 0.2370 - val_acc: 0.9070\n","Epoch 29/30\n","2000/2000 [==============================] - 1s 279us/step - loss: 0.0883 - acc: 0.9760 - val_loss: 0.2376 - val_acc: 0.9070\n","Epoch 30/30\n","2000/2000 [==============================] - 1s 281us/step - loss: 0.0880 - acc: 0.9750 - val_loss: 0.2392 - val_acc: 0.9070\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"O5QFsFj0qc4O","colab_type":"code","colab":{}},"source":["import matplotlib.pyplot as plt"],"execution_count":0,"outputs":[]},{"cell_type":"code","metadata":{"id":"Z5Xb3z_xq6vO","colab_type":"code","outputId":"1b26ce0e-2bd5-40ac-abfd-6c9a7b6b2037","executionInfo":{"status":"ok","timestamp":1580354969360,"user_tz":-540,"elapsed":986,"user":{"displayName":"Yoon Jack","photoUrl":"","userId":"04923927567667044980"}},"colab":{"base_uri":"https://localhost:8080/","height":545}},"source":["acc = history.history['acc']\n","val_acc = history.history['val_acc']\n","loss = history.history['loss']\n","val_loss = history.history['val_loss']\n","\n","epochs = range(len(acc))\n","\n","plt.plot(epochs, acc, 'bo', label='Training acc')\n","plt.plot(epochs, val_acc, 'b', label='Validation acc')\n","plt.title('Training and validation accuracy')\n","plt.legend()\n","\n","plt.figure()\n","\n","plt.plot(epochs, loss, 'bo', label='Training loss')\n","plt.plot(epochs, val_loss, 'b', label='Validation loss')\n","plt.title('Training and validation loss')\n","plt.legend()\n","\n","plt.show()"],"execution_count":13,"outputs":[{"output_type":"display_data","data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAXsAAAEICAYAAAC+iFRkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nIjGgZC8iEgNK9iIiMaBkLyISA/8fkJ4hdoxZfC4AAAAASUVORK5CYII=\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"tags":[]}},{"output_type":"display_data","data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAXUAAAEICAYAAACgQWTXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nKKiLiMSIgrqISIwoqIuIxIiCuohIjPx/TRhTBlhG+p0AAAAASUVORK5CYII=\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"tags":[]}}]},{"cell_type":"code","metadata":{"id":"bEdobQJHq9kP","colab_type":"code","colab":{}},"source":["from keras import models\n","from keras import layers\n","\n","model = models.Sequential()\n","model.add(conv_base)\n","model.add(layers.Flatten())\n","model.add(layers.Dense(256, activation='relu'))\n","model.add(layers.Dense(1, activation='sigmoid'))"],"execution_count":0,"outputs":[]},{"cell_type":"code","metadata":{"id":"pWWZhm5_q_ic","colab_type":"code","outputId":"3dcd2e29-3eab-4f10-d482-cd22203a593d","executionInfo":{"status":"ok","timestamp":1580354971232,"user_tz":-540,"elapsed":345,"user":{"displayName":"Yoon Jack","photoUrl":"","userId":"04923927567667044980"}},"colab":{"base_uri":"https://localhost:8080/","height":305}},"source":["model.summary()"],"execution_count":15,"outputs":[{"output_type":"stream","text":["Model: \"sequential_2\"\n","_________________________________________________________________\n","Layer (type)                 Output Shape              Param #   \n","=================================================================\n","vgg16 (Model)                (None, 4, 4, 512)         14714688  \n","_________________________________________________________________\n","flatten_1 (Flatten)          (None, 8192)              0         \n","_________________________________________________________________\n","dense_3 (Dense)              (None, 256)               2097408   \n","_________________________________________________________________\n","dense_4 (Dense)              (None, 1)                 257       \n","=================================================================\n","Total params: 16,812,353\n","Trainable params: 16,812,353\n","Non-trainable params: 0\n","_________________________________________________________________\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"3VEoEEtSrA1o","colab_type":"code","outputId":"f33ce797-76c0-4fb9-908a-09ee67b75c7c","executionInfo":{"status":"ok","timestamp":1580354972072,"user_tz":-540,"elapsed":334,"user":{"displayName":"Yoon Jack","photoUrl":"","userId":"04923927567667044980"}},"colab":{"base_uri":"https://localhost:8080/","height":35}},"source":["print('conv_base를 동결하기 전 훈련되는 가중치의 수:', \n","      len(model.trainable_weights))"],"execution_count":16,"outputs":[{"output_type":"stream","text":["conv_base를 동결하기 전 훈련되는 가중치의 수: 30\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"KLIpuCXprCvj","colab_type":"code","colab":{}},"source":["conv_base.trainable = False"],"execution_count":0,"outputs":[]},{"cell_type":"code","metadata":{"id":"M8TWaB0xrEEU","colab_type":"code","outputId":"cdb7144c-e19f-43a2-c052-d5e07b30590d","executionInfo":{"status":"ok","timestamp":1580354973593,"user_tz":-540,"elapsed":329,"user":{"displayName":"Yoon Jack","photoUrl":"","userId":"04923927567667044980"}},"colab":{"base_uri":"https://localhost:8080/","height":35}},"source":["print('conv_base를 동결한 후 훈련되는 가중치의 수:', \n","      len(model.trainable_weights))"],"execution_count":18,"outputs":[{"output_type":"stream","text":["conv_base를 동결한 후 훈련되는 가중치의 수: 4\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"zRPDfcKxrFQo","colab_type":"code","outputId":"30f9e176-42cf-4469-f5d3-d8dd1ff61fc9","executionInfo":{"status":"ok","timestamp":1580354974683,"user_tz":-540,"elapsed":730,"user":{"displayName":"Yoon Jack","photoUrl":"","userId":"04923927567667044980"}},"colab":{"base_uri":"https://localhost:8080/","height":53}},"source":["from keras.preprocessing.image import ImageDataGenerator\n","\n","train_datagen = ImageDataGenerator(\n","      rescale=1./255,\n","      rotation_range=20,\n","      width_shift_range=0.1,\n","      height_shift_range=0.1,\n","      shear_range=0.1,\n","      zoom_range=0.1,\n","      horizontal_flip=True,\n","      fill_mode='nearest')\n","\n","# 검증 데이터는 증식되어서는 안 됩니다!\n","test_datagen = ImageDataGenerator(rescale=1./255)\n","\n","train_generator = train_datagen.flow_from_directory(\n","        # 타깃 디렉터리\n","        train_dir,\n","        # 모든 이미지의 크기를 150 × 150로 변경합니다\n","        target_size=(150, 150),\n","        batch_size=20,\n","        # binary_crossentropy 손실을 사용하므로 이진 레이블이 필요합니다\n","        class_mode='binary')\n","\n","validation_generator = test_datagen.flow_from_directory(\n","        validation_dir,\n","        target_size=(150, 150),\n","        batch_size=20,\n","        class_mode='binary')\n","\n","model.compile(loss='binary_crossentropy',\n","              optimizer=optimizers.RMSprop(lr=2e-5),\n","              metrics=['acc'])\n"],"execution_count":19,"outputs":[{"output_type":"stream","text":["Found 2000 images belonging to 2 classes.\n","Found 1000 images belonging to 2 classes.\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"Bqsb_FW6pv-u","colab_type":"code","outputId":"b505cdd8-d725-4f40-c5a9-f20089300131","colab":{"base_uri":"https://localhost:8080/","height":1000},"executionInfo":{"status":"ok","timestamp":1580355566563,"user_tz":-540,"elapsed":591157,"user":{"displayName":"Yoon Jack","photoUrl":"","userId":"04923927567667044980"}}},"source":["history = model.fit_generator(\n","      train_generator,\n","      steps_per_epoch=100,\n","      epochs=30,\n","      validation_data=validation_generator,\n","      validation_steps=50,\n","      verbose=2)"],"execution_count":20,"outputs":[{"output_type":"stream","text":["Epoch 1/30\n"," - 22s - loss: 0.5485 - acc: 0.7320 - val_loss: 0.4090 - val_acc: 0.8430\n","Epoch 2/30\n"," - 20s - loss: 0.4088 - acc: 0.8315 - val_loss: 0.3359 - val_acc: 0.8640\n","Epoch 3/30\n"," - 20s - loss: 0.3496 - acc: 0.8550 - val_loss: 0.2993 - val_acc: 0.8850\n","Epoch 4/30\n"," - 20s - loss: 0.3263 - acc: 0.8695 - val_loss: 0.2916 - val_acc: 0.8850\n","Epoch 5/30\n"," - 20s - loss: 0.2976 - acc: 0.8760 - val_loss: 0.2777 - val_acc: 0.8860\n","Epoch 6/30\n"," - 20s - loss: 0.2896 - acc: 0.8795 - val_loss: 0.2589 - val_acc: 0.9030\n","Epoch 7/30\n"," - 19s - loss: 0.2772 - acc: 0.8865 - val_loss: 0.2561 - val_acc: 0.9040\n","Epoch 8/30\n"," - 20s - loss: 0.2619 - acc: 0.8870 - val_loss: 0.2525 - val_acc: 0.9020\n","Epoch 9/30\n"," - 20s - loss: 0.2615 - acc: 0.8915 - val_loss: 0.2479 - val_acc: 0.8990\n","Epoch 10/30\n"," - 19s - loss: 0.2476 - acc: 0.9035 - val_loss: 0.2438 - val_acc: 0.9040\n","Epoch 11/30\n"," - 20s - loss: 0.2493 - acc: 0.9000 - val_loss: 0.2474 - val_acc: 0.9030\n","Epoch 12/30\n"," - 20s - loss: 0.2309 - acc: 0.9060 - val_loss: 0.2377 - val_acc: 0.9120\n","Epoch 13/30\n"," - 19s - loss: 0.2259 - acc: 0.9105 - val_loss: 0.2358 - val_acc: 0.9130\n","Epoch 14/30\n"," - 20s - loss: 0.2323 - acc: 0.9075 - val_loss: 0.2514 - val_acc: 0.8980\n","Epoch 15/30\n"," - 20s - loss: 0.2099 - acc: 0.9125 - val_loss: 0.2361 - val_acc: 0.9060\n","Epoch 16/30\n"," - 20s - loss: 0.2293 - acc: 0.9050 - val_loss: 0.2356 - val_acc: 0.9100\n","Epoch 17/30\n"," - 20s - loss: 0.2088 - acc: 0.9205 - val_loss: 0.2728 - val_acc: 0.8830\n","Epoch 18/30\n"," - 19s - loss: 0.2183 - acc: 0.9140 - val_loss: 0.2579 - val_acc: 0.8920\n","Epoch 19/30\n"," - 20s - loss: 0.2067 - acc: 0.9130 - val_loss: 0.2600 - val_acc: 0.8940\n","Epoch 20/30\n"," - 20s - loss: 0.2190 - acc: 0.9115 - val_loss: 0.2446 - val_acc: 0.9020\n","Epoch 21/30\n"," - 19s - loss: 0.2028 - acc: 0.9235 - val_loss: 0.2342 - val_acc: 0.9080\n","Epoch 22/30\n"," - 19s - loss: 0.2048 - acc: 0.9200 - val_loss: 0.2360 - val_acc: 0.9080\n","Epoch 23/30\n"," - 19s - loss: 0.1907 - acc: 0.9265 - val_loss: 0.2426 - val_acc: 0.8990\n","Epoch 24/30\n"," - 20s - loss: 0.1926 - acc: 0.9170 - val_loss: 0.2332 - val_acc: 0.9080\n","Epoch 25/30\n"," - 19s - loss: 0.1887 - acc: 0.9235 - val_loss: 0.2325 - val_acc: 0.9090\n","Epoch 26/30\n"," - 19s - loss: 0.1946 - acc: 0.9220 - val_loss: 0.2434 - val_acc: 0.8980\n","Epoch 27/30\n"," - 20s - loss: 0.1911 - acc: 0.9210 - val_loss: 0.2340 - val_acc: 0.9070\n","Epoch 28/30\n"," - 19s - loss: 0.1879 - acc: 0.9220 - val_loss: 0.2442 - val_acc: 0.9030\n","Epoch 29/30\n"," - 19s - loss: 0.1895 - acc: 0.9255 - val_loss: 0.2419 - val_acc: 0.9030\n","Epoch 30/30\n"," - 19s - loss: 0.1874 - acc: 0.9290 - val_loss: 0.2406 - val_acc: 0.9030\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"cXcga2y7rHiu","colab_type":"code","colab":{}},"source":["model.save('/content/drive/My Drive/Colab Notebooks/Keras_creator/model/cats_and_dogs_small_3.h5')"],"execution_count":0,"outputs":[]},{"cell_type":"code","metadata":{"id":"O0rMvh_WrswH","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":545},"outputId":"33457361-1eb2-40b4-dcde-93ef89c35d08","executionInfo":{"status":"ok","timestamp":1580356803413,"user_tz":-540,"elapsed":1025,"user":{"displayName":"Yoon Jack","photoUrl":"","userId":"04923927567667044980"}}},"source":["acc = history.history['acc']\n","val_acc = history.history['val_acc']\n","loss = history.history['loss']\n","val_loss = history.history['val_loss']\n","\n","epochs = range(len(acc))\n","\n","plt.plot(epochs, acc, 'bo', label='Training acc')\n","plt.plot(epochs, val_acc, 'b', label='Validation acc')\n","plt.title('Training and validation accuracy')\n","plt.legend()\n","\n","plt.figure()\n","\n","plt.plot(epochs, loss, 'bo', label='Training loss')\n","plt.plot(epochs, val_loss, 'b', label='Validation loss')\n","plt.title('Training and validation loss')\n","plt.legend()\n","\n","plt.show()"],"execution_count":22,"outputs":[{"output_type":"display_data","data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAYIAAAEICAYAAABS0fM3AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngg==\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"tags":[]}},{"output_type":"display_data","data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAXsAAAEICAYAAAC+iFRkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"tags":[]}}]},{"cell_type":"code","metadata":{"id":"CZRTFsrUruoE","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":845},"outputId":"7a14fead-58e7-4944-8791-7f035285b10a","executionInfo":{"status":"ok","timestamp":1580356804760,"user_tz":-540,"elapsed":552,"user":{"displayName":"Yoon Jack","photoUrl":"","userId":"04923927567667044980"}}},"source":["conv_base.summary()"],"execution_count":23,"outputs":[{"output_type":"stream","text":["Model: \"vgg16\"\n","_________________________________________________________________\n","Layer (type)                 Output Shape              Param #   \n","=================================================================\n","input_1 (InputLayer)         (None, 150, 150, 3)       0         \n","_________________________________________________________________\n","block1_conv1 (Conv2D)        (None, 150, 150, 64)      1792      \n","_________________________________________________________________\n","block1_conv2 (Conv2D)        (None, 150, 150, 64)      36928     \n","_________________________________________________________________\n","block1_pool (MaxPooling2D)   (None, 75, 75, 64)        0         \n","_________________________________________________________________\n","block2_conv1 (Conv2D)        (None, 75, 75, 128)       73856     \n","_________________________________________________________________\n","block2_conv2 (Conv2D)        (None, 75, 75, 128)       147584    \n","_________________________________________________________________\n","block2_pool (MaxPooling2D)   (None, 37, 37, 128)       0         \n","_________________________________________________________________\n","block3_conv1 (Conv2D)        (None, 37, 37, 256)       295168    \n","_________________________________________________________________\n","block3_conv2 (Conv2D)        (None, 37, 37, 256)       590080    \n","_________________________________________________________________\n","block3_conv3 (Conv2D)        (None, 37, 37, 256)       590080    \n","_________________________________________________________________\n","block3_pool (MaxPooling2D)   (None, 18, 18, 256)       0         \n","_________________________________________________________________\n","block4_conv1 (Conv2D)        (None, 18, 18, 512)       1180160   \n","_________________________________________________________________\n","block4_conv2 (Conv2D)        (None, 18, 18, 512)       2359808   \n","_________________________________________________________________\n","block4_conv3 (Conv2D)        (None, 18, 18, 512)       2359808   \n","_________________________________________________________________\n","block4_pool (MaxPooling2D)   (None, 9, 9, 512)         0         \n","_________________________________________________________________\n","block5_conv1 (Conv2D)        (None, 9, 9, 512)         2359808   \n","_________________________________________________________________\n","block5_conv2 (Conv2D)        (None, 9, 9, 512)         2359808   \n","_________________________________________________________________\n","block5_conv3 (Conv2D)        (None, 9, 9, 512)         2359808   \n","_________________________________________________________________\n","block5_pool (MaxPooling2D)   (None, 4, 4, 512)         0         \n","=================================================================\n","Total params: 14,714,688\n","Trainable params: 0\n","Non-trainable params: 14,714,688\n","_________________________________________________________________\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"EUW_6UiBrwja","colab_type":"code","colab":{}},"source":["conv_base.trainable = True\n","\n","set_trainable = False\n","for layer in conv_base.layers:\n","    if layer.name == 'block5_conv1':\n","        set_trainable = True\n","    if set_trainable:\n","        layer.trainable = True\n","    else:\n","        layer.trainable = False"],"execution_count":0,"outputs":[]},{"cell_type":"code","metadata":{"id":"UmkSqQjSryKx","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":1000},"outputId":"c198197e-d0e1-4503-86cb-3cf525e3552a","executionInfo":{"status":"ok","timestamp":1580358943507,"user_tz":-540,"elapsed":2136990,"user":{"displayName":"Yoon Jack","photoUrl":"","userId":"04923927567667044980"}}},"source":["model.compile(loss='binary_crossentropy',\n","              optimizer=optimizers.RMSprop(lr=1e-5),\n","              metrics=['acc'])\n","\n","history = model.fit_generator(\n","      train_generator,\n","      steps_per_epoch=100,\n","      epochs=100,\n","      validation_data=validation_generator,\n","      validation_steps=50)"],"execution_count":25,"outputs":[{"output_type":"stream","text":["Epoch 1/100\n","100/100 [==============================] - 24s 240ms/step - loss: 0.1948 - acc: 0.9215 - val_loss: 0.2266 - val_acc: 0.9110\n","Epoch 2/100\n","100/100 [==============================] - 21s 208ms/step - loss: 0.1717 - acc: 0.9285 - val_loss: 0.2597 - val_acc: 0.8970\n","Epoch 3/100\n","100/100 [==============================] - 21s 212ms/step - loss: 0.1327 - acc: 0.9450 - val_loss: 0.2353 - val_acc: 0.9100\n","Epoch 4/100\n","100/100 [==============================] - 21s 214ms/step - loss: 0.1158 - acc: 0.9545 - val_loss: 0.2144 - val_acc: 0.9200\n","Epoch 5/100\n","100/100 [==============================] - 21s 213ms/step - loss: 0.1084 - acc: 0.9635 - val_loss: 0.2321 - val_acc: 0.9150\n","Epoch 6/100\n","100/100 [==============================] - 22s 217ms/step - loss: 0.0888 - acc: 0.9640 - val_loss: 0.2152 - val_acc: 0.9200\n","Epoch 7/100\n","100/100 [==============================] - 22s 216ms/step - loss: 0.0745 - acc: 0.9695 - val_loss: 0.2253 - val_acc: 0.9270\n","Epoch 8/100\n","100/100 [==============================] - 21s 208ms/step - loss: 0.0695 - acc: 0.9755 - val_loss: 0.2081 - val_acc: 0.9300\n","Epoch 9/100\n","100/100 [==============================] - 21s 214ms/step - loss: 0.0593 - acc: 0.9765 - val_loss: 0.2153 - val_acc: 0.9270\n","Epoch 10/100\n","100/100 [==============================] - 21s 214ms/step - loss: 0.0573 - acc: 0.9780 - val_loss: 0.2059 - val_acc: 0.9290\n","Epoch 11/100\n","100/100 [==============================] - 21s 212ms/step - loss: 0.0568 - acc: 0.9825 - val_loss: 0.2216 - val_acc: 0.9260\n","Epoch 12/100\n","100/100 [==============================] - 21s 214ms/step - loss: 0.0458 - acc: 0.9840 - val_loss: 0.2288 - val_acc: 0.9310\n","Epoch 13/100\n","100/100 [==============================] - 21s 212ms/step - loss: 0.0408 - acc: 0.9865 - val_loss: 0.2265 - val_acc: 0.9240\n","Epoch 14/100\n","100/100 [==============================] - 21s 211ms/step - loss: 0.0452 - acc: 0.9840 - val_loss: 0.2255 - val_acc: 0.9300\n","Epoch 15/100\n","100/100 [==============================] - 22s 218ms/step - loss: 0.0393 - acc: 0.9870 - val_loss: 0.2144 - val_acc: 0.9290\n","Epoch 16/100\n","100/100 [==============================] - 21s 209ms/step - loss: 0.0318 - acc: 0.9885 - val_loss: 0.2249 - val_acc: 0.9310\n","Epoch 17/100\n","100/100 [==============================] - 21s 212ms/step - loss: 0.0256 - acc: 0.9930 - val_loss: 0.2565 - val_acc: 0.9290\n","Epoch 18/100\n","100/100 [==============================] - 22s 216ms/step - loss: 0.0242 - acc: 0.9910 - val_loss: 0.2535 - val_acc: 0.9250\n","Epoch 19/100\n","100/100 [==============================] - 21s 213ms/step - loss: 0.0274 - acc: 0.9905 - val_loss: 0.2444 - val_acc: 0.9300\n","Epoch 20/100\n","100/100 [==============================] - 22s 216ms/step - loss: 0.0275 - acc: 0.9925 - val_loss: 0.2356 - val_acc: 0.9340\n","Epoch 21/100\n","100/100 [==============================] - 22s 216ms/step - loss: 0.0190 - acc: 0.9930 - val_loss: 0.2508 - val_acc: 0.9300\n","Epoch 22/100\n","100/100 [==============================] - 21s 209ms/step - loss: 0.0244 - acc: 0.9920 - val_loss: 0.2459 - val_acc: 0.9310\n","Epoch 23/100\n","100/100 [==============================] - 21s 213ms/step - loss: 0.0196 - acc: 0.9920 - val_loss: 0.3931 - val_acc: 0.9060\n","Epoch 24/100\n","100/100 [==============================] - 21s 212ms/step - loss: 0.0189 - acc: 0.9950 - val_loss: 0.2377 - val_acc: 0.9350\n","Epoch 25/100\n","100/100 [==============================] - 21s 211ms/step - loss: 0.0134 - acc: 0.9945 - val_loss: 0.2687 - val_acc: 0.9250\n","Epoch 26/100\n","100/100 [==============================] - 22s 217ms/step - loss: 0.0205 - acc: 0.9920 - val_loss: 0.2650 - val_acc: 0.9310\n","Epoch 27/100\n","100/100 [==============================] - 21s 211ms/step - loss: 0.0184 - acc: 0.9935 - val_loss: 0.2738 - val_acc: 0.9320\n","Epoch 28/100\n","100/100 [==============================] - 21s 213ms/step - loss: 0.0146 - acc: 0.9955 - val_loss: 0.2578 - val_acc: 0.9320\n","Epoch 29/100\n","100/100 [==============================] - 21s 214ms/step - loss: 0.0131 - acc: 0.9965 - val_loss: 0.3122 - val_acc: 0.9200\n","Epoch 30/100\n","100/100 [==============================] - 21s 211ms/step - loss: 0.0110 - acc: 0.9975 - val_loss: 0.2606 - val_acc: 0.9380\n","Epoch 31/100\n","100/100 [==============================] - 21s 213ms/step - loss: 0.0167 - acc: 0.9940 - val_loss: 0.2812 - val_acc: 0.9330\n","Epoch 32/100\n","100/100 [==============================] - 22s 216ms/step - loss: 0.0110 - acc: 0.9970 - val_loss: 0.2971 - val_acc: 0.9340\n","Epoch 33/100\n","100/100 [==============================] - 21s 210ms/step - loss: 0.0084 - acc: 0.9980 - val_loss: 0.2932 - val_acc: 0.9340\n","Epoch 34/100\n","100/100 [==============================] - 21s 214ms/step - loss: 0.0094 - acc: 0.9975 - val_loss: 0.2913 - val_acc: 0.9280\n","Epoch 35/100\n","100/100 [==============================] - 22s 219ms/step - loss: 0.0123 - acc: 0.9970 - val_loss: 0.2819 - val_acc: 0.9360\n","Epoch 36/100\n","100/100 [==============================] - 21s 212ms/step - loss: 0.0075 - acc: 0.9970 - val_loss: 0.4759 - val_acc: 0.9060\n","Epoch 37/100\n","100/100 [==============================] - 21s 212ms/step - loss: 0.0090 - acc: 0.9980 - val_loss: 0.2993 - val_acc: 0.9370\n","Epoch 38/100\n","100/100 [==============================] - 21s 213ms/step - loss: 0.0069 - acc: 0.9975 - val_loss: 0.2791 - val_acc: 0.9380\n","Epoch 39/100\n","100/100 [==============================] - 21s 207ms/step - loss: 0.0070 - acc: 0.9985 - val_loss: 0.3611 - val_acc: 0.9210\n","Epoch 40/100\n","100/100 [==============================] - 21s 211ms/step - loss: 0.0082 - acc: 0.9970 - val_loss: 0.2605 - val_acc: 0.9410\n","Epoch 41/100\n","100/100 [==============================] - 21s 213ms/step - loss: 0.0070 - acc: 0.9975 - val_loss: 0.3489 - val_acc: 0.9250\n","Epoch 42/100\n","100/100 [==============================] - 21s 208ms/step - loss: 0.0132 - acc: 0.9960 - val_loss: 0.4117 - val_acc: 0.9200\n","Epoch 43/100\n","100/100 [==============================] - 21s 214ms/step - loss: 0.0044 - acc: 0.9990 - val_loss: 0.2698 - val_acc: 0.9420\n","Epoch 44/100\n","100/100 [==============================] - 21s 214ms/step - loss: 0.0063 - acc: 0.9970 - val_loss: 0.3030 - val_acc: 0.9340\n","Epoch 45/100\n","100/100 [==============================] - 21s 214ms/step - loss: 0.0053 - acc: 0.9980 - val_loss: 0.3020 - val_acc: 0.9370\n","Epoch 46/100\n","100/100 [==============================] - 22s 218ms/step - loss: 0.0100 - acc: 0.9970 - val_loss: 0.3038 - val_acc: 0.9360\n","Epoch 47/100\n","100/100 [==============================] - 21s 210ms/step - loss: 0.0060 - acc: 0.9980 - val_loss: 0.3449 - val_acc: 0.9250\n","Epoch 48/100\n","100/100 [==============================] - 21s 212ms/step - loss: 0.0070 - acc: 0.9980 - val_loss: 0.3033 - val_acc: 0.9380\n","Epoch 49/100\n","100/100 [==============================] - 22s 218ms/step - loss: 0.0057 - acc: 0.9970 - val_loss: 0.3498 - val_acc: 0.9180\n","Epoch 50/100\n","100/100 [==============================] - 21s 211ms/step - loss: 0.0028 - acc: 0.9995 - val_loss: 0.3114 - val_acc: 0.9280\n","Epoch 51/100\n","100/100 [==============================] - 21s 211ms/step - loss: 0.0067 - acc: 0.9975 - val_loss: 0.2912 - val_acc: 0.9380\n","Epoch 52/100\n","100/100 [==============================] - 21s 213ms/step - loss: 0.0040 - acc: 0.9990 - val_loss: 0.2979 - val_acc: 0.9330\n","Epoch 53/100\n","100/100 [==============================] - 21s 211ms/step - loss: 0.0047 - acc: 0.9980 - val_loss: 0.3118 - val_acc: 0.9340\n","Epoch 54/100\n","100/100 [==============================] - 21s 213ms/step - loss: 0.0025 - acc: 0.9995 - val_loss: 0.3252 - val_acc: 0.9340\n","Epoch 55/100\n","100/100 [==============================] - 21s 211ms/step - loss: 0.0065 - acc: 0.9975 - val_loss: 0.3177 - val_acc: 0.9350\n","Epoch 56/100\n","100/100 [==============================] - 21s 215ms/step - loss: 0.0064 - acc: 0.9980 - val_loss: 0.3724 - val_acc: 0.9230\n","Epoch 57/100\n","100/100 [==============================] - 21s 212ms/step - loss: 0.0042 - acc: 0.9990 - val_loss: 0.3029 - val_acc: 0.9390\n","Epoch 58/100\n","100/100 [==============================] - 21s 212ms/step - loss: 0.0018 - acc: 0.9995 - val_loss: 0.3303 - val_acc: 0.9350\n","Epoch 59/100\n","100/100 [==============================] - 21s 213ms/step - loss: 0.0026 - acc: 0.9990 - val_loss: 0.3468 - val_acc: 0.9320\n","Epoch 60/100\n","100/100 [==============================] - 22s 220ms/step - loss: 0.0039 - acc: 0.9995 - val_loss: 0.3852 - val_acc: 0.9300\n","Epoch 61/100\n","100/100 [==============================] - 21s 212ms/step - loss: 0.0020 - acc: 0.9995 - val_loss: 0.3284 - val_acc: 0.9330\n","Epoch 62/100\n","100/100 [==============================] - 22s 215ms/step - loss: 0.0028 - acc: 0.9990 - val_loss: 0.3301 - val_acc: 0.9310\n","Epoch 63/100\n","100/100 [==============================] - 21s 215ms/step - loss: 0.0068 - acc: 0.9985 - val_loss: 0.3501 - val_acc: 0.9350\n","Epoch 64/100\n","100/100 [==============================] - 22s 217ms/step - loss: 0.0029 - acc: 0.9995 - val_loss: 0.3621 - val_acc: 0.9300\n","Epoch 65/100\n","100/100 [==============================] - 21s 213ms/step - loss: 0.0047 - acc: 0.9990 - val_loss: 0.3354 - val_acc: 0.9330\n","Epoch 66/100\n","100/100 [==============================] - 21s 214ms/step - loss: 0.0011 - acc: 1.0000 - val_loss: 0.3378 - val_acc: 0.9340\n","Epoch 67/100\n","100/100 [==============================] - 21s 212ms/step - loss: 0.0032 - acc: 0.9990 - val_loss: 0.3945 - val_acc: 0.9300\n","Epoch 68/100\n","100/100 [==============================] - 21s 213ms/step - loss: 0.0062 - acc: 0.9990 - val_loss: 0.3179 - val_acc: 0.9330\n","Epoch 69/100\n","100/100 [==============================] - 21s 214ms/step - loss: 0.0047 - acc: 0.9990 - val_loss: 0.3432 - val_acc: 0.9360\n","Epoch 70/100\n","100/100 [==============================] - 21s 212ms/step - loss: 0.0072 - acc: 0.9985 - val_loss: 0.3110 - val_acc: 0.9350\n","Epoch 71/100\n","100/100 [==============================] - 22s 217ms/step - loss: 0.0055 - acc: 0.9980 - val_loss: 0.3243 - val_acc: 0.9360\n","Epoch 72/100\n","100/100 [==============================] - 21s 210ms/step - loss: 0.0010 - acc: 0.9995 - val_loss: 0.3884 - val_acc: 0.9320\n","Epoch 73/100\n","100/100 [==============================] - 21s 214ms/step - loss: 0.0069 - acc: 0.9975 - val_loss: 0.3567 - val_acc: 0.9290\n","Epoch 74/100\n","100/100 [==============================] - 22s 216ms/step - loss: 6.8281e-04 - acc: 1.0000 - val_loss: 0.3689 - val_acc: 0.9360\n","Epoch 75/100\n","100/100 [==============================] - 21s 215ms/step - loss: 0.0049 - acc: 0.9975 - val_loss: 0.3566 - val_acc: 0.9350\n","Epoch 76/100\n","100/100 [==============================] - 21s 214ms/step - loss: 0.0015 - acc: 0.9995 - val_loss: 0.3731 - val_acc: 0.9330\n","Epoch 77/100\n","100/100 [==============================] - 22s 217ms/step - loss: 0.0033 - acc: 0.9985 - val_loss: 0.3655 - val_acc: 0.9380\n","Epoch 78/100\n","100/100 [==============================] - 21s 212ms/step - loss: 0.0024 - acc: 0.9995 - val_loss: 0.4099 - val_acc: 0.9280\n","Epoch 79/100\n","100/100 [==============================] - 21s 214ms/step - loss: 0.0033 - acc: 0.9980 - val_loss: 0.3844 - val_acc: 0.9350\n","Epoch 80/100\n","100/100 [==============================] - 21s 215ms/step - loss: 0.0023 - acc: 0.9990 - val_loss: 0.3514 - val_acc: 0.9360\n","Epoch 81/100\n","100/100 [==============================] - 21s 209ms/step - loss: 0.0030 - acc: 0.9980 - val_loss: 0.3373 - val_acc: 0.9350\n","Epoch 82/100\n","100/100 [==============================] - 21s 213ms/step - loss: 0.0045 - acc: 0.9995 - val_loss: 0.3446 - val_acc: 0.9390\n","Epoch 83/100\n","100/100 [==============================] - 21s 213ms/step - loss: 0.0044 - acc: 0.9985 - val_loss: 0.3484 - val_acc: 0.9350\n","Epoch 84/100\n","100/100 [==============================] - 21s 212ms/step - loss: 0.0038 - acc: 0.9985 - val_loss: 0.4088 - val_acc: 0.9280\n","Epoch 85/100\n","100/100 [==============================] - 21s 213ms/step - loss: 0.0020 - acc: 0.9995 - val_loss: 0.3724 - val_acc: 0.9340\n","Epoch 86/100\n","100/100 [==============================] - 22s 216ms/step - loss: 0.0010 - acc: 1.0000 - val_loss: 0.4070 - val_acc: 0.9290\n","Epoch 87/100\n","100/100 [==============================] - 21s 209ms/step - loss: 0.0090 - acc: 0.9985 - val_loss: 0.3454 - val_acc: 0.9400\n","Epoch 88/100\n","100/100 [==============================] - 21s 214ms/step - loss: 0.0021 - acc: 0.9990 - val_loss: 0.3489 - val_acc: 0.9390\n","Epoch 89/100\n","100/100 [==============================] - 21s 212ms/step - loss: 0.0042 - acc: 0.9985 - val_loss: 0.5757 - val_acc: 0.9180\n","Epoch 90/100\n","100/100 [==============================] - 21s 214ms/step - loss: 0.0019 - acc: 0.9995 - val_loss: 0.3634 - val_acc: 0.9380\n","Epoch 91/100\n","100/100 [==============================] - 22s 218ms/step - loss: 0.0016 - acc: 0.9990 - val_loss: 0.3365 - val_acc: 0.9350\n","Epoch 92/100\n","100/100 [==============================] - 21s 211ms/step - loss: 6.9939e-04 - acc: 1.0000 - val_loss: 0.3827 - val_acc: 0.9370\n","Epoch 93/100\n","100/100 [==============================] - 22s 216ms/step - loss: 0.0016 - acc: 0.9995 - val_loss: 0.3765 - val_acc: 0.9390\n","Epoch 94/100\n","100/100 [==============================] - 22s 217ms/step - loss: 3.4603e-04 - acc: 1.0000 - val_loss: 0.3657 - val_acc: 0.9390\n","Epoch 95/100\n","100/100 [==============================] - 21s 210ms/step - loss: 2.4160e-04 - acc: 1.0000 - val_loss: 0.3896 - val_acc: 0.9420\n","Epoch 96/100\n","100/100 [==============================] - 22s 216ms/step - loss: 0.0012 - acc: 0.9995 - val_loss: 0.4329 - val_acc: 0.9280\n","Epoch 97/100\n","100/100 [==============================] - 22s 215ms/step - loss: 0.0041 - acc: 0.9985 - val_loss: 0.4421 - val_acc: 0.9380\n","Epoch 98/100\n","100/100 [==============================] - 21s 212ms/step - loss: 0.0046 - acc: 0.9990 - val_loss: 0.4158 - val_acc: 0.9380\n","Epoch 99/100\n","100/100 [==============================] - 22s 215ms/step - loss: 5.2833e-04 - acc: 1.0000 - val_loss: 0.3871 - val_acc: 0.9360\n","Epoch 100/100\n","100/100 [==============================] - 21s 214ms/step - loss: 0.0019 - acc: 0.9990 - val_loss: 0.3537 - val_acc: 0.9410\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"dnRq7WRKr2ys","colab_type":"code","colab":{}},"source":["model.save('/content/drive/My Drive/Colab Notebooks/Keras_creator/model/cats_and_dogs_small_4.h5')"],"execution_count":0,"outputs":[]},{"cell_type":"code","metadata":{"id":"iAoxU-_Yr9ZC","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":545},"outputId":"4e8f1532-4bcb-4ac1-e14b-3b78f6f4318e","executionInfo":{"status":"ok","timestamp":1580358967923,"user_tz":-540,"elapsed":1000,"user":{"displayName":"Yoon Jack","photoUrl":"","userId":"04923927567667044980"}}},"source":["acc = history.history['acc']\n","val_acc = history.history['val_acc']\n","loss = history.history['loss']\n","val_loss = history.history['val_loss']\n","\n","epochs = range(len(acc))\n","\n","plt.plot(epochs, acc, 'bo', label='Training acc')\n","plt.plot(epochs, val_acc, 'b', label='Validation acc')\n","plt.title('Training and validation accuracy')\n","plt.legend()\n","\n","plt.figure()\n","\n","plt.plot(epochs, loss, 'bo', label='Training loss')\n","plt.plot(epochs, val_loss, 'b', label='Validation loss')\n","plt.title('Training and validation loss')\n","plt.legend()\n","\n","plt.show()"],"execution_count":27,"outputs":[{"output_type":"display_data","data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAXoAAAEICAYAAABRSj9aAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAElFTkSuQmCC\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"tags":[]}},{"output_type":"display_data","data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAXQAAAEICAYAAABPgw/pAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nHCdPcEF3HMfJE/4/5ho65ObfAD0AAAAASUVORK5CYII=\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"tags":[]}}]},{"cell_type":"code","metadata":{"id":"XbNduAThr_YI","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":545},"outputId":"e7d54adf-41bf-413c-e060-913df01d907c","executionInfo":{"status":"ok","timestamp":1580358969665,"user_tz":-540,"elapsed":992,"user":{"displayName":"Yoon Jack","photoUrl":"","userId":"04923927567667044980"}}},"source":["def smooth_curve(points, factor=0.8):\n","  smoothed_points = []\n","  for point in points:\n","    if smoothed_points:\n","      previous = smoothed_points[-1]\n","      smoothed_points.append(previous * factor + point * (1 - factor))\n","    else:\n","      smoothed_points.append(point)\n","  return smoothed_points\n","\n","plt.plot(epochs,\n","         smooth_curve(acc), 'bo', label='Smoothed training acc')\n","plt.plot(epochs,\n","         smooth_curve(val_acc), 'b', label='Smoothed validation acc')\n","plt.title('Training and validation accuracy')\n","plt.legend()\n","\n","plt.figure()\n","\n","plt.plot(epochs,\n","         smooth_curve(loss), 'bo', label='Smoothed training loss')\n","plt.plot(epochs,\n","         smooth_curve(val_loss), 'b', label='Smoothed validation loss')\n","plt.title('Training and validation loss')\n","plt.legend()\n","\n","plt.show()"],"execution_count":28,"outputs":[{"output_type":"display_data","data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAXoAAAEICAYAAABRSj9aAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"tags":[]}},{"output_type":"display_data","data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAXoAAAEICAYAAABRSj9aAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nsFL4AAAAAElFTkSuQmCC\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"tags":[]}}]},{"cell_type":"code","metadata":{"id":"K0A0FdMRsBps","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":53},"outputId":"4d5d9c7b-61e8-4b5e-f7e6-54ca88df601e","executionInfo":{"status":"ok","timestamp":1580358974938,"user_tz":-540,"elapsed":4769,"user":{"displayName":"Yoon Jack","photoUrl":"","userId":"04923927567667044980"}}},"source":["test_generator = test_datagen.flow_from_directory(\n","        test_dir,\n","        target_size=(150, 150),\n","        batch_size=20,\n","        class_mode='binary')\n","\n","test_loss, test_acc = model.evaluate_generator(test_generator, steps=50)\n","print('test acc:', test_acc)"],"execution_count":29,"outputs":[{"output_type":"stream","text":["Found 1000 images belonging to 2 classes.\n","test acc: 0.933999993801117\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"3xGrTeP01o_r","colab_type":"code","colab":{}},"source":[""],"execution_count":0,"outputs":[]}]}