starrytong SCNet issues

starrytong / SCNet

MIT License

59 stars 5 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

About evaluation

#23 zhangkexin271 opened 1 week ago
1
Please add these options

#22 dillfrescott closed 4 days ago
3
Fix tuple syntax and tqdm parameters in SCNet.py and apply.py

#21 hunterhogan opened 1 week ago
0
Training behavior with gradient accumulation

#20 lyndonlauder closed 1 week ago
3
The loss and metric suddenly became abnormal during training.

#19 hollandloprabbit opened 2 weeks ago
3
about Model Implementation

#18 autumn-2-net opened 2 weeks ago
6
Please add a config option to save every X steps

#17 dillfrescott closed 2 weeks ago
2
About the training time & training details

#16 LuckyHeart opened 2 months ago
1
Model increase

#15 lucellent opened 2 months ago
4
Large config AMP training: loss=nan, grad=nan, nsdr=nan

#14 ari-ruokamo closed 3 months ago
2
Training other targets than default BVOD?

#13 ari-ruokamo closed 3 months ago
6
About artifact in the silence section

#12 koreyoshi1992 opened 3 months ago
0
Grad=inf and Loss=nan with mixed precision

#11 lyndonlauder closed 3 months ago
3
Further training strategies

#10 ari-ruokamo opened 4 months ago
10
mid and high frequencies missing

#9 fattyyyyyyyyyy opened 4 months ago
6
Update solver.py

#8 starrytong closed 4 months ago
0
Mixed precision error

#7 lyndonlauder closed 4 months ago
1
Intermediate training checkpoint filesizes too large?

#6 ari-ruokamo closed 4 months ago
2
Scnet-large config

#5 256Bab closed 4 months ago
3
Training model with target source

#4 lyndonlauder closed 4 months ago
4
Error inferencing

#3 lucellent closed 4 months ago
3
Training and evaluation code release

#2 lyndonlauder closed 4 months ago
0
Problem with training

#1 ZFTurbo opened 7 months ago
15