in file model.py, _build_var_ema create and var_ema, and apply to all trainable variables, but it didn't use average to update the trainable variables. only in graph_handler.py, when load model from save would use var_ema。
is this a trick?
when training, create shadow variables for trainable variables, but dont update them, when predict, use this to initial the variables .
if it is, then

ub16c9@ub16c9-gpu:/media/ub16c9/fcd84300-9270-4bbd-896a-5e04e79203b7/ub16_prj/myDuReader/preprocess_ml$ python3.6 pre_ml.py data/zhidao/ data/zhidao_82/zhidao.test.json gendata/zhidao_82/zhidao.test sfile/zhidao_82/zhidao.test.result
Traceback (most recent call last):
File "pre_ml.py", line 235, in
main()
File "pre_ml.py", line 231, in main
preprocess(filedir,filename,tar