Lucky Wong

> Any progress? > > Has anyone tried orch version >= 1.9? GPU 1080 ti and TITAN RTX has no problem in PyTorch 1.9.0, but 2080Ti has the same problem.

> Traceback (most recent call last): > File "wenet/bin/train.py", line 199, in > Traceback (most recent call last): > File "wenet/bin/train.py", line 199, in > model, find_unused_parameters=True) > File "/opt/conda/lib/python3.7/site-packages/torch/nn/parallel/distributed.py",...

> > --feedforward-dims "1024,1024,2048,2048,1024" \ > > There are not enough data for aishell1; could you use a smaller model? E.g., `1024,1024,1536,1536,1024` Thanks for the guide, I will try it...

> > --feedforward-dims "1024,1024,2048,2048,1024" \ > > There are not enough data for aishell1; could you use a smaller model? E.g., `1024,1024,1536,1536,1024` Using smaller model, the result is still overfit....

> > Using smaller model, the result is still overfit. Best greedy search result of dev is 5.69。 > > So a smaller model leads to a lower WER. Maybe...

Same commit here: https://github.com/k2-fsa/icefall/pull/962

> https://github.com/jzi040941/PercepNet/blob/8ffae4337d23f920176ac2a7426e84610fa338ab/src/denoise.cpp#L216

> Hey, _stable_solve is a numpy function and works only with numpy arrays. Have you tried it with the torch GPU? Then _stable_solve should fail (Numpy converts CPU tensors to...

`torch_wpe` works fine on GPU device. Test code is here: ```python import numpy as np import torch from nara_wpe import wpe as np_wpe from nara_wpe import torch_wpe T = np.random.randint(100,...

> Thankyou very much for share the code, but when I training a model using the CTCMWER loss, it may be negative, sometime? So what's the reason could be? https://github.com/TeaPoly/CTC-OptimizedLoss/blob/c5b0c17ce5134a6e45fed1cad1a02906932d944d/mwer_loss.py#L75...