Add Azure OpenAI support to RAFT's raft.py #382
Labels
enhancement
New feature or request
Comments
ShishirPatil
pushed a commit
that referenced
this issue
Apr 26, 2024
- Added `--embedding-model` CLI arg defaults to text-embedding-ada-002. Used to generate the chunks embeddings - Added `--completion-model` CLI arg defaults to gpt-4. Used to generate the Q/A dataset - New client_utils.py with two helpers: `build_openai_client` and `build_langchain_embeddings`. - `build_openai_client` builds an `OpenAI` or `AzureOpenAI` client based on detecting `AZURE_*` specific env vars or not - `build_langchain_embeddings` builds a `AzureOpenAIEmbeddings` or `OpenAIEmbeddings` based on detecting `AZURE_*` specific env vars or not - Loads environment variables from .env file using python-dotenv package - Added dependency on python-dotenv - Added doc to README Fixes #382
cedricvidal
changed the title
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Currently, while it is possible to use Azure to run the fine-tuning, it is not currently possible to generate the fine-tuning dataset using Azure OpenAI.
Adding support for running the
raft.pyscript using Azure OpenAI deployments would be very useful for Azure users.Note: doesn't cover
eval.pyThe text was updated successfully, but these errors were encountered: