Merge pull request #5 from nikhilkomakula/rag-gradio-streamlit
Browse files- Dockerfile +0 -1
- README.md +7 -0
- requirements.txt +1 -1
Dockerfile
CHANGED
|
@@ -31,6 +31,5 @@ WORKDIR $HOME/app
|
|
| 31 |
# Clone the Git repo
|
| 32 |
RUN git clone --depth 1 -b deploy-to-hf-spaces https://github.com/nikhilkomakula/llm-rag-op-chatbot.git $HOME/app
|
| 33 |
|
| 34 |
-
|
| 35 |
# Use ENTRYPOINT to specify the command to run when the container starts
|
| 36 |
ENTRYPOINT ["python", "gradio_app.py"]
|
|
|
|
| 31 |
# Clone the Git repo
|
| 32 |
RUN git clone --depth 1 -b deploy-to-hf-spaces https://github.com/nikhilkomakula/llm-rag-op-chatbot.git $HOME/app
|
| 33 |
|
|
|
|
| 34 |
# Use ENTRYPOINT to specify the command to run when the container starts
|
| 35 |
ENTRYPOINT ["python", "gradio_app.py"]
|
README.md
CHANGED
|
@@ -188,6 +188,13 @@ sys.modules['sqlite3'] = sys.modules.pop('pysqlite3')
|
|
| 188 |
|
| 189 |
**Note:** If running locally for Streamlit UI interace and if you hit any errors with `pysqlite3`, try removing whatever that is mentioned above.
|
| 190 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 191 |
## Enhancements:
|
| 192 |
|
| 193 |
* Different advanced retrieval methods could be used.
|
|
|
|
| 188 |
|
| 189 |
**Note:** If running locally for Streamlit UI interace and if you hit any errors with `pysqlite3`, try removing whatever that is mentioned above.
|
| 190 |
|
| 191 |
+
## Hugging Face Docker Space Deployment:
|
| 192 |
+
|
| 193 |
+
* On Hugging Face Docker Space, while installing dependencies, it fails due to dependency conflict. To resolve that, exclude `deepeval` from dependencies and comment out the below listed statements in `eval_rag.py`:
|
| 194 |
+
* `from src.test.eval_custom_model import LLM, eval_rag_metrics`
|
| 195 |
+
* `eval_custom_model = LLM(model_name=EVAL_LLM_NAME, model=hf_eval_llm)`
|
| 196 |
+
* `metrics = eval_rag_metrics(eval_custom_model, question, answer, context)` (set to blank)
|
| 197 |
+
|
| 198 |
## Enhancements:
|
| 199 |
|
| 200 |
* Different advanced retrieval methods could be used.
|
requirements.txt
CHANGED
|
@@ -9,4 +9,4 @@ tensorflow==2.16.1
|
|
| 9 |
gradio==4.21.0
|
| 10 |
pandas==2.2.1
|
| 11 |
streamlit==1.32.2
|
| 12 |
-
|
|
|
|
| 9 |
gradio==4.21.0
|
| 10 |
pandas==2.2.1
|
| 11 |
streamlit==1.32.2
|
| 12 |
+
deepeval==0.21.25
|