{"cells":[{"metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"cell_type":"code","source":"# This Python 3 environment comes with many helpful analytics libraries installed\n# It is defined by the kaggle/python docker image: https://github.com/kaggle/docker-python\n# For example, here's several helpful packages to load in \n\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\nimport matplotlib.pyplot as plt\nimport pydot\nfrom sklearn import preprocessing \nfrom sklearn.model_selection import train_test_split\n\n# Input data files are available in the \"../input/\" directory.\n# For example, running this (by clicking run or pressing Shift+Enter) will list the files in the input directory\n\nimport os\nprint(os.listdir(\"../input\"))\n%matplotlib inline \n\n# Any results you write to the current directory are saved as output.","execution_count":16,"outputs":[{"output_type":"stream","text":"['train.csv', 'sample_submission.csv', 'test.csv']\n","name":"stdout"}]},{"metadata":{},"cell_type":"markdown","source":"## Data import"},{"metadata":{"trusted":true},"cell_type":"code","source":"train = pd.read_csv(\"../input/train.csv\")\noutput=pd.read_csv(\"../input/test.csv\").values\ny=train.label.values\nX=train.drop(['label'],axis=1).values","execution_count":17,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#Visualisation of one image\nsample_index = 8\nplt.figure(figsize=(3, 3))\nplt.imshow(X[sample_index].reshape(28,28), cmap=plt.cm.gray_r, interpolation='nearest')\nplt.title(\"label: %d\" % y[sample_index]);","execution_count":18,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 216x216 with 1 Axes>","image/png":"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\n"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)\n\nX_train_nor = preprocessing.scale(X_train)\nX_test_nor = preprocessing.scale(X_test)\noutput_nor = preprocessing.scale(output)","execution_count":19,"outputs":[{"output_type":"stream","text":"/opt/conda/lib/python3.6/site-packages/sklearn/utils/validation.py:595: DataConversionWarning: Data with input dtype int64 was converted to float64 by the scale function.\n  warnings.warn(msg, DataConversionWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/utils/validation.py:595: DataConversionWarning: Data with input dtype int64 was converted to float64 by the scale function.\n  warnings.warn(msg, DataConversionWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/utils/validation.py:595: DataConversionWarning: Data with input dtype int64 was converted to float64 by the scale function.\n  warnings.warn(msg, DataConversionWarning)\n","name":"stderr"}]},{"metadata":{},"cell_type":"markdown","source":"## Feed Forward Network"},{"metadata":{"trusted":true},"cell_type":"code","source":"from keras.utils.np_utils import to_categorical\n\nY_train = to_categorical(y_train)\nY_test = to_categorical(y_test)","execution_count":20,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### Architecture definition"},{"metadata":{"trusted":true},"cell_type":"code","source":"from keras.models import Sequential\nfrom keras.layers.core import Dense, Activation\n\nN = X_train.shape[1]\nH=128\nK=10\n\nmodel = Sequential()\nmodel.add(Dense(H, input_dim=N,activation=\"relu\"))\nmodel.add(Dense(K, activation=\"softmax\"))\n\nmodel.summary()","execution_count":21,"outputs":[{"output_type":"stream","text":"WARNING:tensorflow:From /opt/conda/lib/python3.6/site-packages/tensorflow/python/framework/op_def_library.py:263: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\nInstructions for updating:\nColocations handled automatically by placer.\n_________________________________________________________________\nLayer (type)                 Output Shape              Param #   \n=================================================================\ndense_1 (Dense)              (None, 128)               100480    \n_________________________________________________________________\ndense_2 (Dense)              (None, 10)                1290      \n=================================================================\nTotal params: 101,770\nTrainable params: 101,770\nNon-trainable params: 0\n_________________________________________________________________\n","name":"stdout"}]},{"metadata":{},"cell_type":"markdown","source":"### Learning"},{"metadata":{"trusted":true},"cell_type":"code","source":"from keras.optimizers import SGD\nsgd = SGD(lr=0.1) ","execution_count":22,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"model.compile(optimizer=sgd,loss='categorical_crossentropy',metrics=['accuracy'])","execution_count":23,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"history = model.fit(X_train_nor, Y_train, epochs=10, batch_size=32,validation_split=0.1)","execution_count":24,"outputs":[{"output_type":"stream","text":"WARNING:tensorflow:From /opt/conda/lib/python3.6/site-packages/tensorflow/python/ops/math_ops.py:3066: to_int32 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.\nInstructions for updating:\nUse tf.cast instead.\nTrain on 30240 samples, validate on 3360 samples\nEpoch 1/10\n30240/30240 [==============================] - 2s 82us/step - loss: 0.3142 - acc: 0.9139 - val_loss: 0.2258 - val_acc: 0.9432\nEpoch 2/10\n30240/30240 [==============================] - 2s 64us/step - loss: 0.1492 - acc: 0.9606 - val_loss: 0.1707 - val_acc: 0.9577\nEpoch 3/10\n30240/30240 [==============================] - 2s 62us/step - loss: 0.1051 - acc: 0.9731 - val_loss: 0.1742 - val_acc: 0.9604\nEpoch 4/10\n30240/30240 [==============================] - 2s 62us/step - loss: 0.0744 - acc: 0.9822 - val_loss: 0.1593 - val_acc: 0.9616\nEpoch 5/10\n30240/30240 [==============================] - 2s 63us/step - loss: 0.0549 - acc: 0.9886 - val_loss: 0.1690 - val_acc: 0.9607\nEpoch 6/10\n30240/30240 [==============================] - 2s 62us/step - loss: 0.0425 - acc: 0.9928 - val_loss: 0.1507 - val_acc: 0.9637\nEpoch 7/10\n30240/30240 [==============================] - 2s 62us/step - loss: 0.0355 - acc: 0.9949 - val_loss: 0.1617 - val_acc: 0.9661\nEpoch 8/10\n30240/30240 [==============================] - 2s 63us/step - loss: 0.0313 - acc: 0.9968 - val_loss: 0.1516 - val_acc: 0.9693\nEpoch 9/10\n30240/30240 [==============================] - 2s 60us/step - loss: 0.0268 - acc: 0.9978 - val_loss: 0.1502 - val_acc: 0.9688\nEpoch 10/10\n30240/30240 [==============================] - 2s 61us/step - loss: 0.0265 - acc: 0.9980 - val_loss: 0.1553 - val_acc: 0.9688\n","name":"stdout"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"plt.figure(figsize=(10,5))\n\nplt.plot(history.history['acc'],label='Train')\nplt.plot(history.history['val_acc'],label='Validation')\nplt.xlabel('# epochs')\nplt.ylabel('Training loss')\n#plt.ylim(0, 6)\nplt.legend(loc='best');","execution_count":25,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 720x360 with 1 Axes>","image/png":"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\n"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"model.evaluate(X_test_nor,Y_test)","execution_count":26,"outputs":[{"output_type":"stream","text":"8400/8400 [==============================] - 0s 33us/step\n","name":"stdout"},{"output_type":"execute_result","execution_count":26,"data":{"text/plain":"[0.17103783099490794, 0.9639285714285715]"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"y_predicted = model.predict_classes(X_test_nor)\nerrors = (y_predicted != y_test)\n\ny_predicted_errors = y_predicted[errors]\ny_test_errors = y_test[errors]\nX_test_errors = X_test[errors]\n\nplt.figure(figsize=(12, 9))\nfor i in range(15):\n    plt.subplot(3, 5, i + 1)\n    plt.imshow(X_test_errors[i].reshape(28, 28),cmap=plt.cm.gray_r, interpolation='nearest')\n    plt.title(\"predicted label: %d\\n true label: %d\"% (y_predicted_errors[i], y_test_errors[i]))\n  ","execution_count":27,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 864x648 with 15 Axes>","image/png":"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\n"},"metadata":{}}]},{"metadata":{},"cell_type":"markdown","source":"## CNN"},{"metadata":{"trusted":true},"cell_type":"code","source":"X_train_nor = X_train_nor.reshape(-1,28,28,1)\nX_test_nor = X_test_nor.reshape(-1,28,28,1)\noutput_nor = output_nor.reshape(-1,28,28,1)","execution_count":28,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"from keras.preprocessing.image import ImageDataGenerator\n\ndatagen = ImageDataGenerator(\n        featurewise_center=False,  # set input mean to 0 over the dataset\n        samplewise_center=False,  # set each sample mean to 0\n        featurewise_std_normalization=False,  # divide inputs by std of the dataset\n        samplewise_std_normalization=False,  # divide each input by its std\n        zca_whitening=False,  # apply ZCA whitening\n        rotation_range=10,  # randomly rotate images in the range (degrees, 0 to 180)\n        zoom_range = 0.1, # Randomly zoom image \n        width_shift_range=0.1,  # randomly shift images horizontally (fraction of total width)\n        height_shift_range=0.1,  # randomly shift images vertically (fraction of total height)\n        horizontal_flip=False,  # randomly flip images\n        vertical_flip=False)  # randomly flip images\n\n\ndatagen.fit(X_train_nor)","execution_count":29,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"datagen.flow(X_train_nor,Y_train,batch_size=10)","execution_count":33,"outputs":[{"output_type":"execute_result","execution_count":33,"data":{"text/plain":"<keras_preprocessing.image.numpy_array_iterator.NumpyArrayIterator at 0x7fda5c469b70>"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"from sklearn.metrics import confusion_matrix\nfrom keras.layers import Dropout, Flatten, Conv2D, MaxPool2D\n\n# Set the CNN model \n# my CNN architechture is In -> [[Conv2D->relu]*2 -> MaxPool2D -> Dropout]*2 -> Flatten -> Dense -> Dropout -> Out\n\nmodel = Sequential()\n\nmodel.add(Conv2D(filters = 32, kernel_size = (5,5),padding = 'Same', activation ='relu', input_shape = (28,28,1)))\nmodel.add(Conv2D(filters = 32, kernel_size = (5,5),padding = 'Same', activation ='relu'))\nmodel.add(MaxPool2D(pool_size=(2,2)))\nmodel.add(Dropout(0.25))\n\n\nmodel.add(Conv2D(filters = 64, kernel_size = (3,3),padding = 'Same', activation ='relu'))\nmodel.add(Conv2D(filters = 64, kernel_size = (3,3),padding = 'Same', activation ='relu'))\nmodel.add(MaxPool2D(pool_size=(2,2), strides=(2,2)))\nmodel.add(Dropout(0.25))\n\n\nmodel.add(Flatten())\nmodel.add(Dense(256, activation = \"relu\"))\nmodel.add(Dropout(0.5))\nmodel.add(Dense(10, activation = \"softmax\"))","execution_count":30,"outputs":[{"output_type":"stream","text":"WARNING:tensorflow:From /opt/conda/lib/python3.6/site-packages/keras/backend/tensorflow_backend.py:3445: calling dropout (from tensorflow.python.ops.nn_ops) with keep_prob is deprecated and will be removed in a future version.\nInstructions for updating:\nPlease use `rate` instead of `keep_prob`. Rate should be set to `rate = 1 - keep_prob`.\n","name":"stdout"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"model.compile(optimizer = sgd , loss = \"categorical_crossentropy\", metrics=[\"accuracy\"])","execution_count":31,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"history = model.fit_generator(datagen.flow(X_train_nor,Y_train,batch_size=64),\n                              epochs=5,\n                              steps_per_epoch = X_train_nor.shape[0] // 64 , \n                              validation_data = (X_test_nor,Y_test))\n","execution_count":41,"outputs":[{"output_type":"stream","text":"Epoch 1/5\n525/525 [==============================] - 154s 293ms/step - loss: 0.6036 - acc: 0.8089 - val_loss: 0.0965 - val_acc: 0.9699\nEpoch 2/5\n525/525 [==============================] - 154s 293ms/step - loss: 0.2103 - acc: 0.9361 - val_loss: 0.0581 - val_acc: 0.9832\nEpoch 3/5\n525/525 [==============================] - 154s 293ms/step - loss: 0.1439 - acc: 0.9557 - val_loss: 0.0475 - val_acc: 0.9848\nEpoch 4/5\n525/525 [==============================] - 153s 292ms/step - loss: 0.1155 - acc: 0.9660 - val_loss: 0.0372 - val_acc: 0.9885\nEpoch 5/5\n525/525 [==============================] - 153s 292ms/step - loss: 0.1061 - acc: 0.9681 - val_loss: 0.0320 - val_acc: 0.9906\n","name":"stdout"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"model.evaluate(X_test_nor,Y_test)","execution_count":42,"outputs":[{"output_type":"stream","text":"8400/8400 [==============================] - 11s 1ms/step\n","name":"stdout"},{"output_type":"execute_result","execution_count":42,"data":{"text/plain":"[0.03196987639281101, 0.9905952380952381]"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"y_predicted = model.predict_classes(X_test_nor)\nerrors = (y_predicted != y_test)\n\ny_predicted_errors = y_predicted[errors]\ny_test_errors = y_test[errors]\nX_test_errors = X_test[errors]\n\nplt.figure(figsize=(24, 18))\nfor i in range(30):\n    plt.subplot(6, 5, i + 1)\n    plt.imshow(X_test_errors[i].reshape(28, 28),cmap=plt.cm.gray_r, interpolation='nearest')\n    plt.title(\"predicted label: %d\\n true label: %d\"% (y_predicted_errors[i], y_test_errors[i]))\n    plt.xticks([])\n  \n  ","execution_count":48,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 1728x1296 with 30 Axes>","image/png":"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\n"},"metadata":{}}]},{"metadata":{},"cell_type":"markdown","source":"## Output creation"},{"metadata":{"trusted":true},"cell_type":"code","source":"y_output = model.predict_classes(output_nor)\ndf_output = pd.DataFrame({\"ImageId\" : np.arange(1,len(y_output)+1),\"Label\": y_output})\ndf_output.set_index(\"ImageId\", inplace=True)\ndf_output.to_csv(\"prediction3.csv\")","execution_count":44,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]}],"metadata":{"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"},"language_info":{"name":"python","version":"3.6.4","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"}},"nbformat":4,"nbformat_minor":1}