OSError: [Errno 12] Cannot allocate memory

[Rock-L]

hi:
when i trained the model , i got the error like:

so i print the system mem every epoch, i find that the available mem decrease every epoch, the process like:

i think every epoch should release the mem , and it's seems not , how can i fix this error? thanks

ps :
dict_core:1000 lines(phrase)
dict_full:2000 lines(phrase)
raw_text:20w lines(words)

[shangjingbo1226]

We will take a look at this. Thanks for pointing it out!

[Rock-L]

We will take a look at this. Thanks for pointing it out!

hi:
have u find something wrong?

[LiyuanLucasLiu]

We did some experiments but didn't reproduce the increase of memory usage (w. our data).

Here are some guesses & TODOs:

1. upgrade the torch-scope to 0.5.1;
2. make sure to use pytorch 1.0;
3. comment the model saving commend during model training;

Best luck to your experiments.