Issues
- 4
audio-visual.pt file
#55 opened by zlwxwxsdzrx - 4
About loss curve
#53 opened by rwbybxx - 1
How AO and VO model are trained?
#54 opened by omerfdeniz - 2
RuntimeError: cublas runtime error
#52 opened by crucian-star - 1
Error happening when evaluating
#51 opened by wcycqjy - 2
preprocess error
#50 opened by ouwen18 - 3
When pretrain VO model, loss reduce a few epochs, but immediately increase and remain.
#43 opened by FL990 - 0
- 1
- 2
The python version 3.6.9 is not supported anymore.
#47 opened by vishwa3344 - 2
Why set "AUDIO_ONLY_PROBABILITY" and "VIDEOO_ONLY_PROBABILITY" these two parameters?
#46 opened by gmh8000 - 1
CUDA out of memory on preprocess.py
#45 opened by InterStella0 - 1
What does <EOS> and " " mean?
#44 opened by FL990 - 5
- 8
- 5
The Validation WER in pretrain is always 1.0
#42 opened by lightning830 - 1
Question about the implement
#41 opened by zero0kiriyu - 9
Transferring to train set was not going well
#37 opened by amadeusuzx - 2
Details of visual fronted model
#39 opened by gmh8000 - 3
ctc_loss problem when running on multiple GPUs
#38 opened by gmh8000 - 1
Transformer Decoder in jointDecoder in avnet
#36 opened by lastlap - 7
WER after pretraining and training
#30 opened by lastlap - 6
The performance when doing curriculum learning
#28 opened by xichenpan - 3
- 1
the result under noisy audio
#27 opened by xjwla - 2
Selection of videos
#26 opened by zhanghousen - 1
- 3
The error when use preprocess.py
#19 opened by xjwla - 0
The error when use preprocess.py
#20 opened by xjwla - 19
Out of Memory errors during pretraining
#3 opened by kasri-mids - 1
How to initialize encoders with AO and VO sets
#18 opened by arunm95 - 2
preprocess for lrs2 dataset
#16 opened by javadpeyman - 7
Pretrained models not available
#8 opened by ashutosh1919 - 1
- 1
The WER is always 1.000
#13 opened by liuzeyang0311 - 3
Curriculum learning
#12 opened by Byshev333 - 3
What does an iteration mean?
#2 opened by kasri-mids - 3
Broken pipe error in Dataloader
#11 opened by smeetrs - 9
- 13
Improvement in AV model performance over AO
#5 opened by smeetrs - 1
- 12
how much training time
#4 opened by yuexianghubit - 6
config.py file is missing?
#1 opened by kasri-mids