def main(): sess = tf.Session() image = read_image('../data/heart.jpg') image = np.reshape(image, [1, 224, 224, 3]) # type numpy.ndarray image.astype(np.float32) parser = Parser('../data/alexnet.cfg') network_builder = NetworkBuilder("test") # type: NetworkBuilder network_builder.set_parser(parser) network = network_builder.build() # type: Network network.add_input_layer(InputLayer(tf.float32, [None, 224, 224, 3])) network.add_output_layer(OutputLayer()) network.connect_each_layer() sess.run(tf.global_variables_initializer()) fc_layer = sess.run(network.output, feed_dict={network.input: image})
def main(): parser = Parser('../data/alexnet.cfg') network_builder = NetworkBuilder("test") mnist = input_data.read_data_sets("F:/tf_net_parser/datasets/MNIST_data/", one_hot=True) # 读取数据 network_builder.set_parser(parser) network = network_builder.build() # type: Network network.add_input_layer(InputLayer(tf.float32, [None, 28, 28, 1])) network.add_output_layer(OutputLayer()) network.set_labels_placeholder(tf.placeholder(tf.float32, [None, 10])) network.connect_each_layer() network.set_accuracy() network.init_optimizer() train_tool = TrainTool() train_tool.bind_network(network) sess = tf.Session() sess.run(tf.initialize_all_variables()) for i in range(300): batch = mnist.train.next_batch(100) feed_dict = { network.input: np.reshape(batch[0], [-1, 28, 28, 1]), network.labels: batch[1] } train_tool.train(sess, network.output, feed_dict=feed_dict) if (i + 1) % 100 == 0: train_tool.print_accuracy(sess, feed_dict) train_tool.save_model_to_pb_file( sess, '../pb/alexnet-' + str(i + 1) + '/', input_data={'input': network.input}, output={'predict-result': network.output}) # train_tool.save_ckpt_model('f:/tf_net_parser/save_model/model', sess, gloabl_step=(i+1)) batch_test = mnist.test.next_batch(100) feed_dict = { network.input: np.reshape(batch_test[0], [100, 28, 28, 1]), network.labels: batch_test[1] } train_tool.print_test_accuracy(sess, feed_dict)