# -*- coding: utf-8 -*-

"""

Created on 2017

@author: 黄文坚、唐源

"""

from datetime import datetime

import math

import time

import tensorflow as tf

def conv_op(input_op, name, kh, kw, n_out, dh, dw, p):

    n_in = input_op.get_shape()[-1].value

    with tf.name_scope(name) as scope:

        #定义卷积层参数：前两个为尺寸 ？*？、第三个为当前层节点矩阵的深度 、第四个为卷积层的深度

        kernel = tf.get_variable(scope+"w",

                                 shape=[kh, kw, n_in, n_out],

                                 dtype=tf.float32,

                                 initializer=tf.contrib.layers.xavier_initializer_conv2d())

        #tf.nn.conv2d 提供了一个方便的卷积层前向传播函数

        #参数1：当前层的节点矩阵，四维矩阵，第一维度对应一个输入batch，如第一张图片，第二张图片..

        #参数2：卷积层参数

        #参数3：不同维度上的步长（第一维、最后一维必须为1）

        #参数4：提供'SAME'和'VALLD'选择，'SAME'为添加全0填充，'VALLD'为不添加

        conv = tf.nn.conv2d(input_op, kernel, (1, dh, dw, 1), padding='SAME')

        #定义偏置项，及下一层节点矩阵的深度 （参数共享）

        bias_init_val = tf.constant(0.0, shape=[n_out], dtype=tf.float32)

        biases = tf.Variable(bias_init_val, trainable=True, name='b')

        #tf.nn.bias_add提供给每个conv节点加上偏置项

        z = tf.nn.bias_add(conv, biases)

        #将计算结果通过ReLU激活函数完成去线性化

        activation = tf.nn.relu(z, name=scope)

        p += [kernel, biases]

        return activation

def fc_op(input_op, name, n_out, p):

    n_in = input_op.get_shape()[-1].value

    with tf.name_scope(name) as scope:

        kernel = tf.get_variable(scope+"w",

                                 shape=[n_in, n_out],

                                 dtype=tf.float32,

                                 initializer=tf.contrib.layers.xavier_initializer())

        biases = tf.Variable(tf.constant(0.1, shape=[n_out], dtype=tf.float32), name='b')

        activation = tf.nn.relu_layer(input_op, kernel, biases, name=scope)

        p += [kernel, biases]

        return activation

def mpool_op(input_op, name, kh, kw, dh, dw):

    #tf.nn.max_pool 提供了一个方便的最大池化层的前向传播过程。

    #tf.nn.avg_pool 提供了一个方便的平均池化层的前向传播过程，两者参数一致。

    #参数1：四维矩阵，第一维度对应一个输入batch，如第一张图片，第二张图片..

    #参数2：ksize为过滤器参数，常为[1, 2, 2, 1]、[1, 3, 3, 1]

    #参数3：不同维度上的步长（第一维、最后一维必须为1）

    #参数4：提供'SAME'和'VALLD'选择，'SAME'为添加全0填充，'VALLD'为不添加

    return tf.nn.max_pool(input_op,

                          ksize=[1, kh, kw, 1],

                          strides=[1, dh, dw, 1],

                          padding='SAME',

                          name=name)

def inference_op(input_op, keep_prob):

    p = []

    # assume input_op shape is 224x224x3

    # block 1 -- outputs 112x112x64

    conv1_1 = conv_op(input_op, name="conv1_1", kh=3, kw=3, n_out=64, dh=1, dw=1, p=p)

    conv1_2 = conv_op(conv1_1,  name="conv1_2", kh=3, kw=3, n_out=64, dh=1, dw=1, p=p)

    pool1 = mpool_op(conv1_2,   name="pool1",   kh=2, kw=2, dw=2, dh=2)

    # block 2 -- outputs 56x56x128

    conv2_1 = conv_op(pool1,    name="conv2_1", kh=3, kw=3, n_out=128, dh=1, dw=1, p=p)

    conv2_2 = conv_op(conv2_1,  name="conv2_2", kh=3, kw=3, n_out=128, dh=1, dw=1, p=p)

    pool2 = mpool_op(conv2_2,   name="pool2",   kh=2, kw=2, dh=2, dw=2)

    # # block 3 -- outputs 28x28x256

    conv3_1 = conv_op(pool2,    name="conv3_1", kh=3, kw=3, n_out=256, dh=1, dw=1, p=p)

    conv3_2 = conv_op(conv3_1,  name="conv3_2", kh=3, kw=3, n_out=256, dh=1, dw=1, p=p)

    conv3_3 = conv_op(conv3_2,  name="conv3_3", kh=3, kw=3, n_out=256, dh=1, dw=1, p=p)

    pool3 = mpool_op(conv3_3,   name="pool3",   kh=2, kw=2, dh=2, dw=2)

    # block 4 -- outputs 14x14x512

    conv4_1 = conv_op(pool3,    name="conv4_1", kh=3, kw=3, n_out=512, dh=1, dw=1, p=p)

    conv4_2 = conv_op(conv4_1,  name="conv4_2", kh=3, kw=3, n_out=512, dh=1, dw=1, p=p)

    conv4_3 = conv_op(conv4_2,  name="conv4_3", kh=3, kw=3, n_out=512, dh=1, dw=1, p=p)

    pool4 = mpool_op(conv4_3,   name="pool4",   kh=2, kw=2, dh=2, dw=2)

    # block 5 -- outputs 7x7x512

    conv5_1 = conv_op(pool4,    name="conv5_1", kh=3, kw=3, n_out=512, dh=1, dw=1, p=p)

    conv5_2 = conv_op(conv5_1,  name="conv5_2", kh=3, kw=3, n_out=512, dh=1, dw=1, p=p)

    conv5_3 = conv_op(conv5_2,  name="conv5_3", kh=3, kw=3, n_out=512, dh=1, dw=1, p=p)

    pool5 = mpool_op(conv5_3,   name="pool5",   kh=2, kw=2, dw=2, dh=2)

    # flatten

    #将第五段卷积网络的输出结果扁平化，转化为 7*7*512=25088的一维向量

    shp = pool5.get_shape()

    flattened_shape = shp[1].value * shp[2].value * shp[3].value

    resh1 = tf.reshape(pool5, [-1, flattened_shape], name="resh1")

    # fully connected

    #连接一个隐含节点为4096的全连接层，激活函数为ReLU,然后连接一个Dropout层

    fc6 = fc_op(resh1, name="fc6", n_out=4096, p=p)

    fc6_drop = tf.nn.dropout(fc6, keep_prob, name="fc6_drop")

    #连接一个隐含节点为4096的全连接层，激活函数为ReLU,然后连接一个Dropout层

    fc7 = fc_op(fc6_drop, name="fc7", n_out=4096, p=p)

    fc7_drop = tf.nn.dropout(fc7, keep_prob, name="fc7_drop")

    #连接一个隐含节点为4096的全连接层，激活函数为ReLU,然后连接一个Softmax进行处理得到分类输出概率

    fc8 = fc_op(fc7_drop, name="fc8", n_out=1000, p=p)

    softmax = tf.nn.softmax(fc8)

    predictions = tf.argmax(softmax, 1)

    return predictions, softmax, fc8, p

def time_tensorflow_run(session, target, feed, info_string):

    num_steps_burn_in = 10

    total_duration = 0.0

    total_duration_squared = 0.0

    for i in range(num_batches + num_steps_burn_in):

        start_time = time.time()

        _ = session.run(target, feed_dict=feed)

        duration = time.time() - start_time

        if i >= num_steps_burn_in:

            if not i % 10:

                print ('%s: step %d, duration = %.3f' %

                       (datetime.now(), i - num_steps_burn_in, duration))

            total_duration += duration

            total_duration_squared += duration * duration

    mn = total_duration / num_batches

    vr = total_duration_squared / num_batches - mn * mn

    sd = math.sqrt(vr)

    print ('%s: %s across %d steps, %.3f +/- %.3f sec / batch' %

           (datetime.now(), info_string, num_batches, mn, sd))

def run_benchmark():

    with tf.Graph().as_default():

        image_size = 224

        images = tf.Variable(tf.random_normal([batch_size,

                                               image_size,

                                               image_size, 3],

                                               dtype=tf.float32,

                                               stddev=1e-1))

        keep_prob = tf.placeholder(tf.float32)

        predictions, softmax, fc8, p = inference_op(images, keep_prob)

        init = tf.global_variables_initializer()

        config = tf.ConfigProto()

        config.gpu_options.allocator_type = 'BFC'

        sess = tf.Session(config=config)

        sess.run(init)

        time_tensorflow_run(sess, predictions, {keep_prob:1.0}, "Forward")

        objective = tf.nn.l2_loss(fc8)

        grad = tf.gradients(objective, p)

        time_tensorflow_run(sess, grad, {keep_prob:0.5}, "Forward-backward")

if __name__ == "__main__":

    batch_size=32

    num_batches=100

    run_benchmark()