Tai Duc Nguyen

Converting GGML Q4_0 back to Torch checkpoint for HuggingFace/Pytorch consumption/training/finetuning

3

Hi everyone, I hacked together a python script to convert a model saved as GGML Q4_0 files back to Pytorch checkpoint for further consumption/training/finetuning using HuggingFace's Transformer package and/or Pytorch/Pytorch...

Converting GGML back to Torch checkpoint for HuggingFace/Pytorch consumption/training/finetuning

5

Here's a PR to convert a model written in a GGML format back to Torch checkpoint for HuggingFace/Pytorch consumption/training/finetuning. Mentioned in issue https://github.com/ggerganov/llama.cpp/issues/359 Also included the ability to use HF's...

Using Torchmetrics with `torch.compile` in Pytorch 2.0

3

## 🐛 Bug I have been trying to get torchmetrics to work smoothly with torch.compile in Pytorch 2.0. However, I got this warning after running through a few training steps:...

### Describe the bug The `extract_compressed_file` and `force_extract` properties of DownloadConfig are always being set to True in the function `dataset_module_factory` in the `load.py` file. This behavior is very annoying...

LitServe MCP Bug: tools/call Always Fails

4

## 🐛 Bug MCP `tools/call` requests consistently fail with `endpoint_handler() missing 1 required positional argument: 'request'` while `tools/list` works perfectly. This makes MCP integration completely unusable. ## Environment - **LitServe...