Reshape 对于的张量x,x.shape=(a, b, c, d)的情况 若调用keras.layer.Reshape(target_shape=(-1, c, d)), 处理后的张量形状为(?, ?, c, d) 若调用tf.reshape(x, shape=[-1, c, d]) 处理后的张量形状为(a*b, c, d) 为了在keras代码中实现tf.reshape的效果,用lambda层做, 调用Lambda(lambda x: tf.reshape(x, shape=[-1, c,