{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"

Khai báo thư viện

\n- Khai báo các thư viện sử dụng
\n- Define các biến toàn cục","metadata":{}},{"cell_type":"code","source":"import os\nimport cv2\nimport tensorflow as tf\nfrom tensorflow import keras\nfrom tensorflow.keras.preprocessing.image import ImageDataGenerator\nfrom tensorflow.keras.layers import Dense, Activation, Dropout, Flatten, BatchNormalization\nfrom tensorflow.keras.layers import Conv2D\nfrom tensorflow.keras.layers import MaxPooling2D\n\nfrom keras.models import load_model\n\nimport numpy as np\nimport timeit\n\nimport matplotlib.pyplot as plt\nfrom matplotlib.image import imread\n\nfrom shutil import copyfile\nfrom shutil import rmtree\n\nimport random\n\nlength = 227\nwidth = 227\nheight = 3\nepoch = 55\nbatchSize = 32\nlearning_rate = 1e-4\ndata_dir = './data'\ncheck_point_path='./checkpoint.h5'","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2021-08-31T11:48:03.634241Z","iopub.execute_input":"2021-08-31T11:48:03.634546Z","iopub.status.idle":"2021-08-31T11:48:08.032361Z","shell.execute_reply.started":"2021-08-31T11:48:03.634479Z","shell.execute_reply":"2021-08-31T11:48:08.03151Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"

Tạo cây thư mục

\nTạo cây thư mục có cấu trúc như bên dưới:\n
\n./data\n├── train\n│   ├── cats\n│   └── dogs\n├── val\n│   ├── cats\n│   └── dogs\n└── test\n    ├── cats\n    └── dogs
","metadata":{}},{"cell_type":"code","source":"def train_validate_test_split(data, train_percent=.6, validate_percent=.2, seed=None):\n np.random.seed(seed)\n perm = np.random.permutation(np.arange(len(data)))\n m = len(data)\n train_end = int(train_percent * m)\n validate_end = int(validate_percent * m) + train_end\n\n train = np.array(data)[:train_end].copy()\n validate = np.array(data)[train_end:validate_end].copy()\n test = np.array(data)[validate_end:-2].copy()\n\n return train, validate, test\n\n\nif os.path.isdir(data_dir):\n print('Deleting data directory...',end='')\n rmtree(data_dir)\n print('Done!',end='\\n')\n\nos.makedirs(data_dir, exist_ok=True)\n\ndata_dir = data_dir + '/'\n\nsubdirs = ['train/', 'val/', 'test/']\n\nprint('Create subdirectory.......',end='')\nfor subdir in subdirs:\n # Create label subdirectories\n labeldirs = ['dogs/', 'cats/']\n\n for labldir in labeldirs:\n newdir = data_dir + subdir + labldir\n os.makedirs(newdir, exist_ok=True)\n\nprint('Done!',end='\\n')\n\n# Copy dataset into folders\ndog_files = []\ncat_files = []\n\n# Copy training dataset images into subdirectories\nsrc_directory = '../input/dogs-vs-cats/dogs-vs-cats/train/train'\n\nfor file in os.listdir(src_directory):\n if file.startswith('cat'):\n cat_files.append(file)\n elif file.startswith('dog'):\n dog_files.append(file)\n\nprint('Cat:', len(cat_files))\nprint('Dog:', len(dog_files))\n\ntrain_cats, val_cats, test_cats = train_validate_test_split(\n cat_files, seed=42, train_percent=.6, validate_percent=.2)\ntrain_dogs, val_dogs, test_dogs = train_validate_test_split(\n dog_files, seed=42, train_percent=.6, validate_percent=.2)\n\nprint('Classify directory........',end ='')\nlist(map(lambda file: copyfile(src_directory + '/' + file, data_dir + 'train/cats/' + file), train_cats))\nlist(map(lambda file: copyfile(src_directory + '/' + file, data_dir + 'train/dogs/' + file), train_dogs))\nlist(map(lambda file: copyfile(src_directory + '/' + file, data_dir + 'val/cats/' + file), val_cats))\nlist(map(lambda file: copyfile(src_directory + '/' + file, data_dir + 'val/dogs/' + file), val_dogs))\nlist(map(lambda file: copyfile(src_directory + '/' + file, data_dir + 'test/cats/' + file), test_cats))\nlist(map(lambda file: copyfile(src_directory + '/' + file, data_dir + 'test/dogs/' + file), test_dogs))\n\nprint(\"Done!\",end='\\n')","metadata":{"execution":{"iopub.status.busy":"2021-08-31T11:48:08.033884Z","iopub.execute_input":"2021-08-31T11:48:08.034208Z","iopub.status.idle":"2021-08-31T11:51:07.847877Z","shell.execute_reply.started":"2021-08-31T11:48:08.034174Z","shell.execute_reply":"2021-08-31T11:51:07.846994Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"

Data Augment

\nSử dụng ImageDataGenerator để tạo thêm các ảnh mới từ tập ảnh gốc bằng cách: lật ngang, lật dọc, kéo méo ảnh, phóng to, thu nhỏ ảnh, tăng giảm sáng, tương phản, thêm noise…
\n→ Nhằm đa dạng góc nhìn về các bức ảnh để train model","metadata":{}},{"cell_type":"code","source":"# Generating images for the Test set\nprint('Generating images.........', end='')\ntrain_datagen = ImageDataGenerator(rescale=1.0/255.0,\n width_shift_range=0.1, height_shift_range=0.1, horizontal_flip=True)\n\nval_datagen = ImageDataGenerator(rescale=1.0/255.0)\ntest_datagen = ImageDataGenerator(rescale=1.0/255.0)\n\n# Creating training set\ntrain_it = train_datagen.flow_from_directory(data_dir +'/train/',\n class_mode='binary', batch_size=batchSize, target_size=(length, width))\nval_it = val_datagen.flow_from_directory(data_dir +'/val/',\n class_mode='binary', batch_size=batchSize, target_size=(length, width))\n\n# Creating the Test set\ntest_it = test_datagen.flow_from_directory(data_dir +'/test/',\n class_mode='binary', batch_size=batchSize, target_size=(length, width))\n\nprint('Done!',end='\\n')","metadata":{"execution":{"iopub.status.busy":"2021-08-31T11:51:07.849852Z","iopub.execute_input":"2021-08-31T11:51:07.85022Z","iopub.status.idle":"2021-08-31T11:51:09.17752Z","shell.execute_reply.started":"2021-08-31T11:51:07.850181Z","shell.execute_reply":"2021-08-31T11:51:09.176719Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"

Hàm Define CNN model

\nDùng mô hình AlexNet
\n* Reference: AlexNet Architecture
\n* Reference: Cats and Dogs classification using AlexNet ","metadata":{}},{"cell_type":"code","source":"def optimizer_init_fn():\n return tf.keras.optimizers.Adam(learning_rate)\n\ndef defineModel():\n print('Defining model..........',end='')\n # initializing the CNN\n model = keras.Sequential()\n\n model.add(Conv2D(64, (3, 3), activation=\"relu\", padding=\"same\",\n kernel_initializer='he_uniform', name='conv1', input_shape=(length,width,height)))\n\n model.add(Conv2D(96, (11,11), strides= 4,\n padding= 'valid', activation= 'relu',\n input_shape= (length,width,height),\n kernel_initializer= 'he_normal', name='conv2'))\n model.add(MaxPooling2D(pool_size=(3,3), strides= (2,2),\n padding= 'valid', data_format= None, name='maxpool1'))\n\n model.add(Conv2D(256, (5,5), strides= 1,\n padding= 'same', activation= 'relu',\n kernel_initializer= 'he_normal', name='conv3'))\n model.add(MaxPooling2D(pool_size=(3,3), strides= (2,2),\n padding= 'valid', data_format= None, name='maxpool2')) \n\n model.add(Conv2D(384, (3,3), strides= 1,\n padding= 'same', activation= 'relu',\n kernel_initializer= 'he_normal', name='conv4'))\n\n model.add(Conv2D(384, (3,3), strides= 1,\n padding= 'same', activation= 'relu',\n kernel_initializer= 'he_normal', name='conv5'))\n\n model.add(Conv2D(256, (3,3), strides= 1,\n padding= 'same', activation= 'relu',\n kernel_initializer= 'he_normal', name='conv6'))\n\n model.add(MaxPooling2D(pool_size=(3,3), strides= (2,2),\n padding= 'valid', data_format= None, name='maxpool3'))\n\n model.add(Flatten())\n model.add(Dense(4096, activation= 'relu'))\n model.add(Dense(4096, activation= 'relu'))\n model.add(Dense(1000, activation= 'relu'))\n model.add(Dense(1, activation= 'sigmoid'))\n\n # compile model\n model.compile(optimizer=optimizer_init_fn(),\n loss='binary_crossentropy',\n metrics=['accuracy']) \n\n print('Done!',end='\\n')\n return model","metadata":{"execution":{"iopub.status.busy":"2021-08-31T11:51:09.180355Z","iopub.execute_input":"2021-08-31T11:51:09.180622Z","iopub.status.idle":"2021-08-31T11:51:09.194384Z","shell.execute_reply.started":"2021-08-31T11:51:09.180596Z","shell.execute_reply":"2021-08-31T11:51:09.193296Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"

Load model

\n- Nạp model đang được train dở từ trước
\n- File model được train dở: ./checkpoint.h5\n","metadata":{}},{"cell_type":"code","source":"# kiểm tra nếu model đã train dở từ trước thì tiến hành nạp và chạy\nif os.path.exists(check_point_path) == True:\n print('Model loading....',end='')\n # nạp model đã chạy dở từ lần trước\n model = load_model(check_point_path)\n print('Done!',end='\\n')\nelse:\n print('Model had not trained before')","metadata":{"execution":{"iopub.status.busy":"2021-08-31T11:51:09.195693Z","iopub.execute_input":"2021-08-31T11:51:09.196203Z","iopub.status.idle":"2021-08-31T11:51:09.209982Z","shell.execute_reply.started":"2021-08-31T11:51:09.196169Z","shell.execute_reply":"2021-08-31T11:51:09.209181Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"

Callback

\n* Reference: Callback
\nLưu lại các trạng thái và thông số của model trong quá trình training vào file checkpoint.hdf5
\n- Stop quá trình training khi model đạt được một kết quả accuracy/ loss nhất định nào đó
\n- Lưu lại model sau mỗi epoch
\n\n","metadata":{}},{"cell_type":"code","source":"earlyStopping = tf.keras.callbacks.EarlyStopping(monitor='loss', patience=5, restore_best_weights=True)\ncheckPoint = tf.keras.callbacks.ModelCheckpoint(check_point_path, monitor='val_accuracy', verbose=1, save_best_only=True, mode='auto')\n\ncallback = [earlyStopping, checkPoint]","metadata":{"execution":{"iopub.status.busy":"2021-08-31T11:51:09.211256Z","iopub.execute_input":"2021-08-31T11:51:09.211607Z","iopub.status.idle":"2021-08-31T11:51:09.218452Z","shell.execute_reply.started":"2021-08-31T11:51:09.211559Z","shell.execute_reply":"2021-08-31T11:51:09.21759Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"

Hàm Train model

\n","metadata":{}},{"cell_type":"code","source":"def trainModel(model,train_it, val_it,epoch):\n history = model.fit_generator(train_it, steps_per_epoch=len(train_it),\n validation_data=val_it, validation_steps=len(val_it), epochs=epoch, verbose=1,\n callbacks=[callback])\n\n return history","metadata":{"_kg_hide-input":false,"_kg_hide-output":false,"execution":{"iopub.status.busy":"2021-08-31T11:51:09.219718Z","iopub.execute_input":"2021-08-31T11:51:09.22013Z","iopub.status.idle":"2021-08-31T11:51:09.227366Z","shell.execute_reply.started":"2021-08-31T11:51:09.220098Z","shell.execute_reply":"2021-08-31T11:51:09.226651Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"

Hàm Lưu model đã train

\n* Reference: Save and load model","metadata":{}},{"cell_type":"code","source":"def saveModel(file_name):\n print('Saving model.......',end='')\n model.save('final_model.h5')\n print('Done!', end='\\n')","metadata":{"execution":{"iopub.status.busy":"2021-08-31T11:51:09.230541Z","iopub.execute_input":"2021-08-31T11:51:09.231428Z","iopub.status.idle":"2021-08-31T11:51:09.237684Z","shell.execute_reply.started":"2021-08-31T11:51:09.231385Z","shell.execute_reply":"2021-08-31T11:51:09.236849Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"

Hàm Đánh giá mô hình

\n- Sau khi train xong, tiến hành đánh giá qua tập data test_it và in ra độ chính xác của phân loại
\n- Địa chỉ của tập test_it: ./data/test\n","metadata":{}},{"cell_type":"code","source":"def evaluateModel(test_it):\n # evaluate model\n print('Evaluate model.......',end='')\n _, acc = model.evaluate_generator(test_it, steps=len(test_it), verbose=0)\n print('> accuracy: %.3f' % (acc * 100.0), end='\\n') ","metadata":{"execution":{"iopub.status.busy":"2021-08-31T11:51:09.241015Z","iopub.execute_input":"2021-08-31T11:51:09.241302Z","iopub.status.idle":"2021-08-31T11:51:09.247761Z","shell.execute_reply.started":"2021-08-31T11:51:09.241279Z","shell.execute_reply":"2021-08-31T11:51:09.246953Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"

Hàm Visualize Training

\n- Vẽ đồ thị Visualize quá trình training
\n- Tóm tắt mô hình ","metadata":{}},{"cell_type":"code","source":"def summarize_diagnostics(history):\n plt.plot(history.history['accuracy'])\n plt.plot(history.history['val_accuracy'])\n plt.title('model accuracy')\n plt.ylabel('accuracy')\n plt.xlabel('epoch')\n plt.legend(['train', 'val'], loc='upper left')\n plt.show()\n\n plt.plot(history.history['loss'])\n plt.plot(history.history['val_loss'])\n plt.title('model loss')\n plt.ylabel('loss')\n plt.xlabel('epoch')\n plt.legend(['train', 'val'], loc='upper left')\n plt.show()","metadata":{"execution":{"iopub.status.busy":"2021-08-31T11:51:09.251008Z","iopub.execute_input":"2021-08-31T11:51:09.251314Z","iopub.status.idle":"2021-08-31T11:51:09.257524Z","shell.execute_reply.started":"2021-08-31T11:51:09.251283Z","shell.execute_reply":"2021-08-31T11:51:09.256745Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"

Hàm Test dự đoán kết quả

\n- Tiến hành dự đoán kết quả từ model đã train
\n- Model đạt độ chính xác: 93.375%","metadata":{}},{"cell_type":"code","source":"def predict(model, link):\n image = cv2.imread(link)\n image = cv2.resize(image, (length,width))\n preds = model.predict(np.expand_dims(image, axis=0))[0]\n plt.imshow(image)\n plt.show()\n if preds==0:\n print(\"Predicted:Cat\")\n else:\n print(\"Predicted: Dog\") \n \ndef test(model,test_dir, numtest):\n print('Testing.........')\n test_cat = test_dir + '/cats'\n test_dog = test_dir + '/dogs'\n for i in range(numtest):\n # test cat\n filename = random.choice(os.listdir(test_cat))\n predict(model, test_cat + '/' + filename) \n \n # test dog\n filename = random.choice(os.listdir(test_dog))\n predict(model, test_dog + '/' + filename) ","metadata":{"execution":{"iopub.status.busy":"2021-08-31T11:51:09.259135Z","iopub.execute_input":"2021-08-31T11:51:09.259656Z","iopub.status.idle":"2021-08-31T11:51:09.272768Z","shell.execute_reply.started":"2021-08-31T11:51:09.259613Z","shell.execute_reply":"2021-08-31T11:51:09.271915Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"

(Main) Test model đã train từ trước

\n- Các ảnh test được đặt trong mục ./data/test
\n- Hàm này lấy ngẫu nhiên numtest chó, mèo ra test","metadata":{}},{"cell_type":"code","source":"print('Model loading.......',end='')\nmodel = load_model('../input/dogs-vs-cats/final_model.h5')\nprint('Done!',end='\\n')\n\nevaluateModel(test_it)\n\nmodel.summary()\n\ntest(model,'./data/test',3)","metadata":{"execution":{"iopub.status.busy":"2021-08-31T11:51:09.273832Z","iopub.execute_input":"2021-08-31T11:51:09.274201Z","iopub.status.idle":"2021-08-31T11:51:46.122124Z","shell.execute_reply.started":"2021-08-31T11:51:09.274165Z","shell.execute_reply":"2021-08-31T11:51:46.121282Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"

(Main) Train và chạy lại từ đầu

","metadata":{}},{"cell_type":"code","source":"model = defineModel()\n# kiểm tra nếu model đã train dở từ trước thì tiến hành nạp và chạy\nif os.path.exists(check_point_path) == True:\n print('Model loading......Done!')\n # nạp model đã chạy dở từ lần trước\n model = load_model(check_point_path)\nelse:\n print('Model had not trained before')\n \n \nhistory = trainModel(model,train_it, val_it, epoch)\n\nsaveModel('final_model.h5')\n\nevaluateModel(test_it)\n\nsummarize_diagnostics(history)\n\nmodel.summary()\n\ntest(model,'./data/test',3)","metadata":{"execution":{"iopub.status.busy":"2021-08-31T11:51:46.123507Z","iopub.execute_input":"2021-08-31T11:51:46.123877Z","iopub.status.idle":"2021-08-31T12:57:34.863234Z","shell.execute_reply.started":"2021-08-31T11:51:46.123838Z","shell.execute_reply":"2021-08-31T12:57:34.860024Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"

Kết quả tóm tắt lịch sử train

\nĐây là lịch sử sau khi train 55 epoch\n\n\n![model_accuracy.png](attachment:342f83c3-20bc-4cde-985c-929f9131dffb.png)\n![model_loss.png](attachment:57b0d3a1-f29f-4a9f-814e-2867e3f92732.png)","metadata":{},"attachments":{"342f83c3-20bc-4cde-985c-929f9131dffb.png":{"image/png":"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"},"57b0d3a1-f29f-4a9f-814e-2867e3f92732.png":{"image/png":"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"}}}]}