dynamic_prompting
1.0.0
与pip:
pip install dynamic_prompting
或从源安装
git clone https://github.com/ElmiraGhorbani/dynamic_prompting
pip install -e .
访问Meta Llama网站(https://llama.meta.com/llama-downloads)并注册以下载型号/s。
注册后,您将收到带有URL的电子邮件以下载模型。运行下载脚本时,您将需要此URL。
收到电子邮件后,请导航到下载的Llama存储库并运行下载。SH脚本。
下载模型后,将它们放入此文件夹中。
否则您可以使用这种方式:
- Step 1: Go to your Hugging Face account “Settings” and then “Access Tokens” on the left column, and copy the token you need.
- Step 2: On your terminal, export your token starting with “HF_”. Use a distinct name (for example, HF_TOKEN) for each token you export.
You may add this line to your ~/.bashrc if you do not want to export it every time you start a session.
export HF_TOKEN="HF_XXXXXXXXXXXXX"
嵌入式类设计用于与文本嵌入模型接口。有许多嵌入模型提供商(OpenAi,Cohere,拥抱的脸等)。目前,该类提供了MXBAI EMBED-LARGE-V1 , BGE-SMALL-EN-V1.5和NOMIC-EMBED-TEXT-V1.5的标准接口。
首先,下载模型并将其放入此文件夹中。
git lfs install
git clone https://huggingface.co/nomic-ai/nomic-embed-text-v1.5
cd nomic-embed-text-v1.5
git lfs fetch
git clone https://huggingface.co/BAAI/bge-small-en-v1.5
cd bge-small-en-v1.5
git lfs fetch
git clone https://huggingface.co/mixedbread-ai/mxbai-embed-large-v1
cd mxbai-embed-large-v1
git lfs fetch
检查示例。IPYNB