when training the SACN on WN18RR, CUDA is our of memory(The batch size is 64)

JD-AI-Research-Silicon-Valley / SACN

End-to-end Structure-Aware Convolutional Networks for Knowledge Base Completion

MIT License

113 stars 30 forks source link

Open learntolearn11 opened 4 years ago

maqy1995 commented 4 years ago

try torch.cuda.empty_cache() after every batch may help free about 2GB memory(in my experiment).

chaoshangcs commented 4 years ago

@maqy1995 Thanks for your help to response this question.