Permission error on model llama2_7B

Models
1

I am trying to download LLAMA2_7B Model on local network.
So I used huggingface - files and versions and got these files into local network.
I used a code by vs code and used
[python convert_llama_weights_to_hf.py --input_dir Z:\Documents\Download\7B --model_size 7B --output_dir model]
and this error occured.

image
image1559×755 29 KB

the last sentence is meaning by
[PermissionError: [WinError 32] The process cannot access the file because it is being used by another process]
Can you tell me how to solve this problem?
Or can you recommend me other better solutions to run llama on local networks?

2

Hi,

Could you clarify why you’re running the conversion script? It’s recommended to just load the HF weights right away as shown in the model card: meta-llama/Llama-2-7b-hf · Hugging Face.

If you need them in a local network, then load the model as shown in the model card, call save_pretrained on it to save the weights in a local folder, which you can then pass around. One can use the from_pretrained method to load the weights from the folder.

4

Thanks for your help, I solved this problem.