{"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 tensorflow as tf\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\nfrom sklearn.model_selection import train_test_split\ntf.logging.set_verbosity('INFO')\n\nimport os\nprint(os.listdir(\"../input\"))\n\n# Any results you write to the current directory are saved as output.","execution_count":null,"outputs":[]},{"metadata":{"_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","trusted":true,"_kg_hide-output":true},"cell_type":"code","source":"train = pd.read_csv('../input/train.csv')\ntest = pd.read_csv('../input/test.csv')\n\n# print(train.shape)\ntrain_label = train.pop('label')\n\nprint(train.info())\nprint(test.info())\nprint(train_label.value_counts())","execution_count":null,"outputs":[]},{"metadata":{"_kg_hide-output":true,"_kg_hide-input":true,"trusted":true,"_uuid":"a55bbff5e36a6140809f0d7b83251c9f048f6e06"},"cell_type":"code","source":"#Normalization\ntrain /= 255\ntest /= 255","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"c74e291f50be78ab39699096bdc0769ee03df39e"},"cell_type":"code","source":"#split data to train and validation set\ntrain_features, val_features, train_label, val_label = train_test_split(\n    train, train_label,\n    train_size=0.75\n)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"7e3dc94c87056173c6cc5b0bf483236577ea6b2e"},"cell_type":"code","source":"#prepare input functions\ntrain_input_fn = tf.estimator.inputs.numpy_input_fn(\n    x = {'x' : train_features.values},\n    y = train_label.values,\n    num_epochs=None,\n    shuffle=True\n)\nval_input_fn = tf.estimator.inputs.numpy_input_fn(\n    x = {'x' : val_features.values},\n    y = val_label.values,\n    shuffle=False\n)\npredict_input_fn = tf.estimator.inputs.numpy_input_fn(x = {'x' : test.values}, shuffle=False)\n","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"ab9c00e3dce261cb95db83121525c37d369b8d57"},"cell_type":"code","source":"def model_fn(features, labels, mode):\n\n    '''\n    this mode contains :\n    conv -> max_pool -> conv -> max_pool->dense->dropout->softmax\n\n    :param features: input features, has only one key 'x'\n    :param labels:  with shape [num_samples, 10]\n    :param mode: TRAIN EVAL PREDICT\n    :return: EstimatorSpec\n    '''\n    inputs = tf.reshape(features['x'], [-1, 28,28,1])\n    conv1 = tf.layers.conv2d(\n        inputs = inputs,\n        filters=32,\n        kernel_size=[5,5],\n        padding='same',\n        activation=tf.nn.relu\n    )\n    pool1 = tf.layers.max_pooling2d(\n        inputs=conv1,\n        pool_size=[2,2],\n        strides=2\n    )\n    conv2 = tf.layers.conv2d(\n        inputs = pool1,\n        filters=64,\n        kernel_size=[5,5],\n        padding='same',\n        activation=tf.nn.relu\n    )\n    pool2 = tf.layers.max_pooling2d(\n        inputs=conv2,\n        pool_size=[2,2],\n        strides=2\n    )\n    pool2 = tf.reshape(pool2, [-1, 7*7*64])\n    dense = tf.layers.dense(\n        inputs=pool2,\n        units=1024,\n        activation=tf.nn.relu\n    )\n    dropout = tf.layers.dropout(\n        inputs = dense,\n        training = mode == tf.estimator.ModeKeys.TRAIN\n    )\n    logits = tf.layers.dense(inputs=dropout, units=10)\n\n    predicts = {\n        'classid' : tf.argmax(logits, axis=1),\n        'probabilities' : tf.nn.softmax(logits)\n    }\n    if mode == tf.estimator.ModeKeys.PREDICT:\n        return tf.estimator.EstimatorSpec(mode=mode, predictions=predicts)\n\n    losses = tf.losses.sparse_softmax_cross_entropy(labels=labels, logits=logits)\n    if mode == tf.estimator.ModeKeys.TRAIN:\n        optimizer = tf.train.GradientDescentOptimizer(0.01)\n        train_op = optimizer.minimize(losses, global_step=tf.train.get_global_step())\n        return tf.estimator.EstimatorSpec(mode, loss=losses, train_op=train_op)\n\n    eval_metrics = {'accuracy' : tf.metrics.accuracy(labels, predicts['classid'])}\n    return tf.estimator.EstimatorSpec(mode, loss=losses, eval_metric_ops=eval_metrics)\n","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"202c4bc946f2817e8e042d534a872a5451c3ddcd"},"cell_type":"code","source":"#create estimator\ncnn_classifier = tf.estimator.Estimator(\n    model_fn=model_fn,\n    model_dir='mode'\n)\n","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"2b0dda1c907ed1645d2979d2913321843cf46167"},"cell_type":"code","source":"#train and eval\ncnn_classifier.train(train_input_fn, steps=1000)\neval_result = cnn_classifier.evaluate(val_input_fn)\nprint('eval accuracy is {}'.format(eval_result['accuracy']))\npredict_result = cnn_classifier.predict(predict_input_fn)","execution_count":null,"outputs":[]}],"metadata":{"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"},"language_info":{"name":"python","version":"3.6.6","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"}},"nbformat":4,"nbformat_minor":1}