Thorin Farnsworth

I had this issue with `run_speech_recognition_ctc.py` for wa2vec2.0 fine-tuning. I made a small change and the hash for the function (which includes tokenisation) is now the same before and after...

`tokenizer.save_pretrained(training_args.output_dir)` produces a different tokenizer hash when loaded on restart of the script. When I was debugging before I was terminating the script prior to this command, then rerunning. I...

I've ported wav2vec2.0 fine-tuning into Optimum-Graphcore which is where I found the issue. The majority of the script was copied from the Transformers version to keep it similar, [here is...

its only applied to half of the feature dimension. [here](https://www.kaggle.com/code/aeryss/rotary-postional-encoding-rope-pytorch) is another implementation from a kaggle post: ```python import torch class Rotary(torch.nn.Module): def __init__(self, dim, base=10000): super().__init__() inv_freq = 1....

[The stochastic sampler](https://github.com/openai/consistency_models/blob/6d26080c58244555c031dbc63080c0961af74200/cm/karras_diffusion.py#L657) is the same, assuming you finish on `t_min`. In this case the scaling applied to the noise will be zero, so it finishes with the final denoised...

Been a while since I've been in depth with this code, so this may be a naive question. Are you sure `ts` should be in ascending order? Having the larger...

The ordering is different in the code compared to the algorithm in paper, but they amount to the same thing: ```python t = (t_max_rho + ts[0] / (steps - 1)...

> > > Been a while since I've been in depth with this code, so this may be a naive question. Are you sure `ts` should be in ascending order?...

Make sure your commands aren't starting with `mpiexec -n 8` like some of the scripts suggest. This is a multi-GPU command.

Which command are you using? I'll take a look