mpt-7B-inference/download_model.py

import os
from huggingface_hub import hf_hub_download


def download_mpt_quant(destination_folder: str, repo_id: str, model_filename: str):
    local_path = os.path.abspath(destination_folder)
    return hf_hub_download(
        repo_id=repo_id,
        filename=model_filename,
        local_dir=local_path,
        local_dir_use_symlinks=True
    )


if __name__ == "__main__":
    """full url: https://huggingface.co/TheBloke/mpt-30B-chat-GGML/blob/main/mpt-30b-chat.ggmlv0.q4_1.bin"""

    repo_id = "TheBloke/mpt-30B-chat-GGML"
    model_filename = "mpt-30b-chat.ggmlv0.q4_1.bin"
    destination_folder = "models"
    download_mpt_quant(destination_folder, repo_id, model_filename)
initial commit 2023-06-26 05:36:27 +00:00			`import os`
			`from huggingface_hub import hf_hub_download`


cleanup code 2023-06-26 07:10:40 +00:00			`def download_mpt_quant(destination_folder: str, repo_id: str, model_filename: str):`
use absolute path 2023-06-26 15:16:46 +00:00			`local_path = os.path.abspath(destination_folder)`
initial commit 2023-06-26 05:36:27 +00:00			`return hf_hub_download(`
cleanup code 2023-06-26 07:10:40 +00:00			`repo_id=repo_id,`
			`filename=model_filename,`
remove cache dir 2023-06-26 17:23:01 +00:00			`local_dir=local_path,`
use symlink download, caches to the hf cache folder instead -> avoids downloading model twice if download is run again 2023-06-26 18:05:03 +00:00			`local_dir_use_symlinks=True`
initial commit 2023-06-26 05:36:27 +00:00			`)`


			`if __name__ == "__main__":`
			`"""full url: https://huggingface.co/TheBloke/mpt-30B-chat-GGML/blob/main/mpt-30b-chat.ggmlv0.q4_1.bin"""`

cleanup code 2023-06-26 07:10:40 +00:00			`repo_id = "TheBloke/mpt-30B-chat-GGML"`
			`model_filename = "mpt-30b-chat.ggmlv0.q4_1.bin"`
Fix typo 2023-06-26 07:50:51 +00:00			`destination_folder = "models"`
cleanup code 2023-06-26 07:10:40 +00:00			`download_mpt_quant(destination_folder, repo_id, model_filename)`