有没有能补一下的谢谢

# coding=utf-8
import numpy as np
import struct
import os
import time

class FullyConnectedLayer(object):
    def __init__(self, num_input, num_output):  # 全连接层初始化
        self.num_input = num_input
        self.num_output = num_output
        print('\tFully connected layer with input %d, output %d.' % (self.num_input, self.num_output))
    def init_param(self, std=0.01):  # 参数初始化
        self.weight = np.random.normal(loc=0.0, scale=std, size=(self.num_input, self.num_output))
        self.bias = np.zeros([1, self.num_output])
    def forward(self, input):  # 前向传播计算
        start_time = time.time()
        self.input = input
        # TODO：全连接层的前向传播，计算输出结果
        self.output = _______________
        return self.output
    def backward(self, top_diff):  # 反向传播的计算
        # TODO：全连接层的反向传播，计算参数梯度和本层损失
        self.d_weight = ________________
        self.d_bias = ________________
        bottom_diff = ________________
        return bottom_diff
    def update_param(self, lr):  # 参数更新
        # TODO：对全连接层参数利用参数进行更新
        self.weight = ________________
        self.bias = ________________
    def load_param(self, weight, bias):  # 参数加载
        assert self.weight.shape == weight.shape
        assert self.bias.shape == bias.shape
        self.weight = weight
        self.bias = bias
    def save_param(self):  # 参数保存
        return self.weight, self.bias

class ReLULayer(object):
    def __init__(self):
        print('\tReLU layer.')
    def forward(self, input):  # 前向传播的计算
        start_time = time.time()
        self.input = input
        # TODO：ReLU层的前向传播，计算输出结果
        output = ________________
        return output
    def backward(self, top_diff):  # 反向传播的计算
        # TODO：ReLU层的反向传播，计算本层损失
        bottom_diff = ________________
        return bottom_diff

class SoftmaxLossLayer(object):
    def __init__(self):
        print('\tSoftmax loss layer.')
    def forward(self, input):  # 前向传播的计算
        # TODO：softmax 损失层的前向传播，计算输出结果
        input_max = np.max(input, axis=1, keepdims=True)
        input_exp = np.exp