Issues
- 0
- 0
How to train a model to separate an instrument
#43 opened by gstr001 - 0
[Torch 1.13.1>2.0.0) RuntimeError: istft requires a complex-valued input tensor matching the output from stft with return_complex=True
#42 opened by deton24 - 2
A Baseline for SDX 2023
#41 opened by ws-choi - 1
It is possible to separate a music track with this model on a computer with an Intel CPU without an Nvidia GPU, with or without OpenVINO
#40 opened by insinfo - 8
- 8
- 0
How do you save the mixer model?
#38 opened by KimberleyJensen - 0
- 1
- 2
Separation without training
#34 opened by kichel98 - 1
Error on building pesq "pesq/cypesq.c:6:10: fatal error: Python.h: No such file or directory"
#32 opened by cnh2769 - 2
- 0
TFC-TDF-U-Net's performance on Musdb18
#31 opened by ws-choi - 0
Data Parallel
#26 opened by rlaalstjr47 - 2
Final model configs
#23 opened by rlaalstjr47 - 0
cmd, wandb train loss is different
#19 opened by rlaalstjr47 - 2
Data Augmentation
#6 opened by ws-choi - 0
initialization
#8 opened by ws-choi - 0
quick fixes
#17 opened by ws-choi - 0
Sync Batchnorm test
#11 opened by ws-choi - 0
Yaml Integration
#12 opened by ws-choi - 0
auto lr find
#13 opened by ws-choi - 1
duplicated wandb session created
#9 opened by rlaalstjr47 - 2
wandb only logs top k val loss
#15 opened by rlaalstjr47 - 1
- 0
yaml for 12, 13 servers
#7 opened by ws-choi - 0
Validation for DDP
#4 opened by ws-choi - 3
mskim revision
#3 opened by ws-choi