apply_model run time
mathpopo opened this issue · 5 comments
mathpopo commented
i try to use separate(htdemuc),but i meet : i put 100ms / 1s ,apply_model run time is same,how can i do to see different?
CarlGao4 commented
What system are you using? What is your CPU and GPU?
mathpopo commented
ubuntu 20.04, GPU, other issue:all model point:
stereo,44100?
i put a mono 16000, but ->torch audio load : stereo,44100
mathpopo commented
padded_mix = mix.padded(valid_length), should here again enlange the length, end set segment "1", just 44100
so:
- 16k mono->stereo,44100
- length->padded
because segment,so every epoch is same time
CarlGao4 commented
What? Are you TRAINING demucs?