代码来源:https://github.com/eriklindernoren/ML-From-Scratch
卷积神经网络中卷积层Conv2D(带stride、padding)的具体实现:https://cloud.tencent.com/developer/article/1686529
激活函数的实现(sigmoid、softmax、tanh、relu、leakyrelu、elu、selu、softplus):https://cloud.tencent.com/developer/article/1686496
损失函数定义(均方误差、交叉熵损失):https://cloud.tencent.com/developer/article/1686498
优化器的实现(SGD、Nesterov、Adagrad、Adadelta、RMSprop、Adam):https://cloud.tencent.com/developer/article/1686499
卷积层反向传播过程:https://cloud.tencent.com/developer/article/1686503
全连接层实现:https://cloud.tencent.com/developer/article/1686504
批量归一化层实现:https://cloud.tencent.com/developer/article/1686506
池化层实现:https://cloud.tencent.com/developer/article/1686507
padding2D实现:https://cloud.tencent.com/developer/article/1686509
这就相当于是pytorch中的在全连接层之前使用view()函数类似的操作:
class Flatten(Layer):
""" Turns a multidimensional matrix into two-dimensional """
def __init__(self, input_shape=None):
self.prev_shape = None
self.trainable = True
self.input_shape = input_shape
def forward_pass(self, X, training=True):
self.prev_shape = X.shape
return X.reshape((X.shape[0], -1))
def backward_pass(self, accum_grad):
return accum_grad.reshape(self.prev_shape)
def output_shape(self):
return (np.prod(self.input_shape),)需要注意反向传播时的形状的改变。
还有Reshape层:
class Reshape(Layer):
""" Reshapes the input tensor into specified shape
Parameters:
-----------
shape: tuple
The shape which the input shall be reshaped to.
"""
def __init__(self, shape, input_shape=None):
self.prev_shape = None
self.trainable = True
self.shape = shape
self.input_shape = input_shape
def forward_pass(self, X, training=True):
self.prev_shape = X.shape
return X.reshape((X.shape[0], ) + self.shape)
def backward_pass(self, accum_grad):
return accum_grad.reshape(self.prev_shape)
def output_shape(self):
return self.shape