读取保存的TFmodel出现NotFoundError

NotFoundError: Unsuccessful TensorSliceReader constructor: Failed to find any matching files for ../target-predict/data/kegg_GPopt_model/2018-05-15-16-02-01/67, [1478], 0.0005691415809378206, 0.0004931419580688316, 0.27912254612734133, 40.pickle

解决：https://github.com/tensorflow/tensorflow/issues/6082 

Could you try the following:

1. use a model name without the character [], and:
2. when you tried to restore, use the full relative path ./model_epoch10 rather than model_epoch10

If my guess is right then you should see it work. It it doesn't, could you give the steps so that people can reproduce the problem (i.e. how to use your code).

总结：

TF的model的名字不能带有 [ ] , : 这三个字符

于是把最好的模型复制出来重命名了，解决。