EvoNet icon indicating copy to clipboard operation
EvoNet copied to clipboard

ValueError: Input graph and Layer graph are not the same: Tensor("Reshape_1:0", shape=(?, 256), dtype=float32) is not from the passed-in graph.

Open zouyihui opened this issue 2 years ago • 1 comments

您好,我想请教一下,当我执行net = tf.layers.dense(patterns, 3, activation=tf.nn.relu, name="outnet_fc1")的时候报错: ValueError: Input graph and Layer graph are not the same: Tensor("Reshape_1:0", shape=(?, 256), dtype=float32) is not from the passed-in graph.可能是什么问题,应该怎么解决了

zouyihui avatar Jun 07 '22 05:06 zouyihui

这是 输入的张量shape 不对,检查下数据输入

VachelHU avatar Jun 07 '22 05:06 VachelHU