# 1. TFLearn API definitions winder neural network. Import tflearn Import tflearn.datasets.mnist AS MNIST from tflearn.layers.conv Import conv_2d, max_pool_2d from tflearn.layers.estimator Import Regression from tflearn.layers.core Import Input_Data, Dropout, fully_connected trainX, trainY, testX, testy = MNIST. load_data (data_dir = " F.: 201806-GitHub \\ \\ \\ TensorFlowGoogle Datasets MNIST_data \\ " , one_hot = True) # to resize the image data into a format convolution neural network input. = trainX.reshape trainX ([-. 1, 28, 28,. 1 ]) testX= testX.reshape([-1, 28, 28, 1]) # 构建神经网络。 net = input_data(shape=[None, 28, 28, 1], name='input') net = conv_2d(net, 32, 5, activation='relu') net = max_pool_2d(net, 2) net = conv_2d(net, 64, 5, activation='relu') net = max_pool_2d(net, 2) net = fully_connected(net, 500, activation='relu') net = fully_connected(net, 10, activation='softmax') # Define learning task. SGD designated optimizer, learning of 0.01, the loss of cross-entropy function. Regression = NET (NET, Optimizer = ' SGD ' , learning_rate = 0.01, Loss = ' categorical_crossentropy ' )
# 2. train the neural network by TFLearn the API. # A network architecture defined training model, the effect of the model validation and verification on the data specified. = tflearn.DNN Model (NET, tensorboard_verbose = 0) model.fit (trainX, trainY, n_epoch = 10, validation_set = ([testX, testy]), show_metric = True)
