Skip to content

Instantly share code, notes, and snippets.

@iphysresearch
Last active February 11, 2018 10:55
Show Gist options
  • Save iphysresearch/4f0006723cf186195e5685c74408e398 to your computer and use it in GitHub Desktop.
Save iphysresearch/4f0006723cf186195e5685c74408e398 to your computer and use it in GitHub Desktop.
test
def affine_forward(x, w, b):
"""
Inputs:
- x: A numpy array containing input data, of shape (N, d_1, ..., d_k) 样本
- w: A numpy array of weights, of shape (D, M) 权重
- b: A numpy array of biases, of shape (M,) 偏置
Returns a tuple of:
- out: output, of shape (N, M)
- cache: (x, w, b)
"""
out = None # 初始化
reshaped_x = np.reshape(x, (x.shape[0],-1)) # 确保x是一个规整的矩阵
out = reshape_x.dot(w) +b # out = w x +b
cache = (x, w, b)
# cache将该函数的输入值缓冲储存起来,以备后面计算梯度时使用
return out, cache
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment