Tensorflow XLA编译错误：操作没有Atft命名&γx27；XlaCompile'；

我正在运行下面的代码。我正在获得XLA编译器错误，如下所述。我有Tensorflow GPU 1.8版本，但我尝试了CPU版本，我得到了错误。但是当我将Tensorflow（CPU）降级到1.3时，我没有得到任何错误。Tensorflow是使用带有Pip的虚拟环境安装的

有人能帮我一下吗

代码：

错误日志：

(?, 16, 16, 3)
(?, 8, 8, 6)
(?, 16, 16, 3)     # <<<<< This should have printed (?, ?, ?, ?)

Traceback (most recent call last):
  File "/my/path/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 2327, in get_attr
    c_api.TF_OperationGetAttrValueProto(self._c_op, name, buf)
tensorflow.python.framework.errors_impl.InvalidArgumentError: Operation 'conv2d_transpose' has no attr named '_XlaCompile'.

Traceback (most recent call last):
  File "/my/path/lib/python3.6/site-packages/tensorflow/python/ops/gradients_impl.py", line 380, in _MaybeCompile
    xla_compile = op.get_attr("_XlaCompile")
  File "/my/path/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 2331, in get_attr
    raise ValueError(str(e))
ValueError: Operation 'conv2d_transpose' has no attr named '_XlaCompile'.

（？，16，16，3）
(?, 8, 8, 6)
（？，16，16，3）发生问题的原因是过滤器的形状不匹配以及与之相关的步幅，以产生所需的输出形状和有效的填充。我应该将反褶积滤波器的形状设置为[1,1,3,3]，并将步长设置为[1,1,1,1]

所需代码：

deconv_filter = tf.get_variable('deconv_filter', shape=[1, 1, 6, 3], dtype=tf.float32)

deconv = tf.nn.conv2d_transpose(input_tensor, filter=deconv_filter,
    output_shape=tf.shape(input_tensor),
    strides=[1, 1, 1, 1],
    padding='VALID')

deconv_filter = tf.get_variable('deconv_filter', shape=[1, 1, 6, 3], dtype=tf.float32)

deconv = tf.nn.conv2d_transpose(input_tensor, filter=deconv_filter,
    output_shape=tf.shape(input_tensor),
    strides=[1, 1, 1, 1],
    padding='VALID')