-
Using released llmcompressor 0.1.0 on python 3.11 on ubuntu 20.04
Phi3Small Instruct does not have the default weights in the mapping (q_proj, k_proj, v_proj), so I supplied my own and it failed wi…
-
This issue is to gather research on color contrast. If you have links to color contrast research or are aware of individuals working in this space, please note that here.
Peer reviewed research s…
-
我使用doc中的方法进行pretrain和Finetune 过程
生成的文件目录为
```
-rw-r--r-- 1 work work 11000 Sep 25 08:10 adapter_config.json
-rw-r--r-- 1 work work 323020440 Sep 25 08:10 adapter_model.safetensors
-rw-r--r-…
-
Just like that title says, can we use microsoft/Phi-3-vision-128k-instruct-onnx-cpu to run on mobile in similar way we are running https://huggingface.co/microsoft/Phi-3-mini-4k-instruct-onnx?
-
Right now `/v1` is part of ENDPOINT.
But it is [more common to have BASE_URL](https://github.com/openai/openai-python?tab=readme-ov-file#configuring-the-http-client) (or API_BASE) setting, which incl…
-
```text
BadRequestError: Error code: 400 - {'object': 'error', 'message': 'Attempted to assign 1 x 2509 = 2509 image tokens to 0 placeholders', 'type': 'BadRequestError', 'param': None, 'code': 400…
-
## Description
When I try to run Phi-3-vision with flash attention using DJL serving on sagemaker (0.29.0 image), I get a warning saying "Cannot use FlashAttention-2 backend due to sliding window."…
-
I'm trying to run the following code in kaggle with **GPUP100**
`!bash /kaggle/working/Phi3-Vision-Finetune/scripts/finetune_lora_vision.sh`
### complete error
`[2024-09-14 09:33:24,960] [INFO] …
-
### Your current environment
```text
Collecting environment information...
PyTorch version: 2.3.0+cu121
Is debug build: False
CUDA used to build PyTorch: 12.1
ROCM used to build PyTorch: N/A
…
-
https://huggingface.co/Qwen/Qwen-VL-Chat/tree/main
https://huggingface.co/deepseek-ai/deepseek-vl-7b-chat
I've gotten extremely good results off of these, would be great to have them baseline in…