Tensorflow AutoEncoder：当前的实现还不支持批量和深度维度上的步幅

Question

我正在尝试创建一个自动编码器解码器框架。下面是我使用的代码

# Define convolution layers
def conv(layer_name, input_X, shape, strides, padding = "SAME"):
    with tf.variable_scope(layer_name):
        W = tf.get_variable("W", shape = shape, dtype=tf.float32)
        return tf.nn.conv2d(input_X, W, strides, padding), W

# Layer1 convolution
encoder_layer1, W1 = conv("encode_layer1", X, [28, 28, 1, 10], [2,2,2,2])

我收到以下错误

InvalidArgumentError (see above for traceback): Current implementation does not yet support strides in the batch and depth dimensions.
     [[Node: encode_layer1/Conv2D = Conv2D[T=DT_FLOAT, data_format="NHWC", dilations=[1, 1, 1, 1], padding="SAME", strides=[2, 2, 2, 2], use_cudnn_on_gpu=true, _device="/job:localhost/replica:0/task:0/device:CPU:0"](_arg_X_0_0, encode_layer1/W/read)]]

Answer 1

从 Tensorflow documentation 看到这个：

Must have strides[0] = strides[3] = 1. For the most common case of the same horizontal and vertices strides, strides = [1, stride, stride, 1].

strides[0] 是批量维度，strides[3] 是通道（或深度）维度。

尝试将步幅设置为 [1, 2, 2, 1]。

Tensorflow AutoEncoder：当前的实现还不支持批量和深度维度上的步幅

Tensorflow AutoEncoder: Current implementation does not yet support strides in the batch and depth dimensions

python

autoencoder

tensorflow