Yashiro1016

The ["Loss nan"](https://github.com/YiwenShaoStephen/pychain_example/issues/6) issue says that samples having too few #frames will cause this error (similar to the case in CTC where input_length < target_length). This problem occurs when I...

Hello, I dug deeper into the code and found the tensors became nan after passing the TDNN(1D dilation layer) The input tensors still have values before going into the TDNN...

Hello, I actually got an error message during step 0 (although the loss is inf, not nan)  I got this error message right after the model structure is printed...

Hello, After several trial & errors, I decided to decrease the size of the dataset to speed up my debugging process (from 111656 to 19874) And...the bug just disappeared itself,...

Hello, I did some experiments and found that the system crashes when training data reaches about 80k~90k utterances. If the amount of data reached this number, the system will crash...