search

maum-ai / voicefilter

Unofficial PyTorch implementation of Google AI's VoiceFilter system
http://swpark.me/voicefilter
1.09k stars 227 forks source link

use torch.no_grad scope when inferencing #6

Closed seungwonpark closed 5 years ago

seungwonpark commented 5 years ago

This reduces the usage of GPU memory when inferencing about half.

Missing torch.no_grad() scope in inference.py was my mistake.