s1162276945
s1162276945
> return torch.eq(y_preds,y_test_var).data.numpy().sum()/x_test_var.size(0) Excuse me, but what is the reason? I have changed the file as you say, but still show "Accuracy of the model tensor(0)" maybe it's the problem...
> why accuracy of the model tensor(0)? >  I have met the same problem with you.
> @s1162276945 , I am not maintaining the repo anymore. The code was tested in pytorch v0.3, looks like much changes are done to pytorch. Try running the code in...
in python3, you currently have to import dill as pickle https://stackoverflow.com/questions/25348532/can-python-pickle-lambda-functions
Loading data... Load word2vec u file /home/wsy/keras/data-sharing/GoogleNews-vectors-negative300.bin header: b'3000000 300\n' ---------------------------运行到这里就停止不动了------------------------- 机器：CentOS Tesla K40m
Loading data... 2018-05-30 14:41:01.384097: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.1 instructions, but these are available on your machine and could speed up CPU computations. 2018-05-30 14:41:01.384169:...
我换了小一点的预训练数据集SLIM，还是会卡住。
> 可能你的运行环境没有显卡？ 是的，所以我用cpu跑着测试一下
> 可以在trainer.train()后面加上对于model的保存操作 加上了，但是加载的时候出错了，可以请教一下你么
> > 可以在trainer.train()后面加上对于model的保存操作 > > 加上了，但是加载的时候出错了，可以请教一下你么 > > 已经发邮件给你了，还请帮忙看一下，谢谢。