Yuekai Zhang

I would like take on this. - [ ] Support the Onnxruntime CUDA provider.

Yes, even we do noting with H, it would cause the wrong output of torch.nonzero()

> Can you try to trace back the error as early as possible? I assume the output of np.load is being affected, e.g. data[ctc_log_probs] is being affected. You might be...

> It says > > ``` > ========= This tool is deprecated and will be removed in a future release of the CUDA toolkit > ========= Please use the compute-sanitizer...

I put the ctc_topo call after the np.load related codes. This time it without print hang anymore, however, reports k2 errors: [swipe.txt](https://github.com/k2-fsa/k2/files/9036376/swipe.txt). Also, I tested the code, it works on...

> @yuekaizhang can you try with different values of max_token and see if it depends on that? You may be able to do some kind of bisection to find out...

> It might be something to do with the device properties. What device? I am using v100 32gb.

I would like try Pruned_Stateless_Transducer_2 on aishell2 if no one is doing it.

Hi, Dan, my comparison is based on the code: https://github.com/k2-fsa/snowfall/blob/2dda31e14039a79b77c89bcd3bb96d52cbf60c8a/snowfall/training/ctc_graph.py#L108-L127. I used the below code to avoid composing operation. I was wondering any reference code to compile them as a...

Thanks. I could try this way to compose. Actually, for my code, I follow this requirement _When treat_epsilons_specially is True, this function works only on CPU. When treat_epsilons_specially is False...