graphvite
graphvite copied to clipboard
Memory OOM saving model.
It costs about 130G memory training model but when to save model it occurs oom error(total memory 195G). Is there extra memory allocation when saving the model?
There shouldn't be large memory allocation. Maybe it is caused by pickle?
We will check the memory usage later.
There shouldn't be large memory allocation. Maybe it is caused by
pickle?We will check the memory usage later.
It is caused by pickle. I replace joblib with pickle. The joblib uses less memory when dumps model but it costs more disk capacity.