Suhas

I am trying to prune transformer model, but after pruning the model results in layer mismatch `mat1 and mat2 shapes cannot be multiplied (86x635 and 0x768)` Any reasons for this,...

I am following examples of transformer pruning, But if i just increase the amount more than 10%. Model accuracy drops more, any suggestion to maintain the accuracy??

After distillation, the logs of training a student model is very large. you can find about model here https://huggingface.co/deepset/gelectra-large i am following distillation as following https://kd-lib.readthedocs.io/en/latest/usage/tutorials/VanillaKD.html ``` Epoch: 10, Loss:...

Computing loss within onnxrunitme inference (GPT2 model)

4

This could be a feature, i am trying to compute language model metries. i.e., PERPLEXITY score. In pytorch it is convenient to get score by ``` tokenizer = GPT2Tokenizer.from_pretrained('gpt2') model...

Mt5 model loading fails

11

Hallo, I have MT5 pretrained model, i am using fastt5 approch to convert the model to onnx. The convestion of the model works fine. But when creating the decoder_sess at...

MT5 model convertion with sparseml

8

Hello, I am trying to converting mt5 model to onnx runtime. But i can see that model is not supported. Is there any work around. Error: `Unrecognized configuration class for...

Example for custom models

1

The framework show really good results, is there any examples for custom models distilation process.

Mlflow logs multiple times in distributed training

7

Hallo, I have been training model in distributed pytorch using hugging face trainer API. Now i have been training model on slrum multi node multi gpu and for every GPU,...