Skip to content

RbKomar/-winners-LLM-Dec-Hackathon

Repository files navigation

Demo Files Overview

This section provides guidance on running demonstration files located in the knowledge_graph directory of the legacy_code_assistant project.

Directory Path

  • Location: .demo/

Demo Files

  • github_graph_DEMO_1_GIF.py: Visualizes GitHub graphs.

  • repo_code_graph_DEMO_2_GIF.py: Demonstrates repository code graphs.

    • Note for DEMO 2:
      • To integrate an LLM into this demo, locate the TODO comment in the code where additional processing should be added.
      • Currently, the template selection feature is not operational. This needs to be fixed to enable proper functionality.

Running Streamlit Demos

To run the Streamlit demonstrations for each script, use the following commands in PowerShell:

  • For GitHub Graph Visualization:
    streamlit run github_graph_DEMO_1_GIF.py
  • For Repository Code Graph:
    streamlit run repo_code_graph_DEMO_2_GIF.py

Repository Setup

Ensure the repository path is set to tests/test_repo. To clone the required repository, execute the repo_cloner script:

python legacy_code_assistant/data_extraction/repo_cloner.py

Credentials

You must provide credentials in the demo folder. A file credentials.yaml should be placed there. It should have the following structure:

'AZURE_OPENAI_ENDPOINT': <URL OF AZURE ENDPOINT>
'AZURE_OPENAI_API_KEY': <KEY>
'Deployment_completion': <AZURE DEPLOYMENT FOR COMPLETION MODEL (e.g. gpt4)>
'Deployment_embeddings':  <AZURE DEPLOYMENT FOR EMBEDDING MODEL (e.g. ada)>

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 4

  •  
  •  
  •  
  •