Skip to content

JTpath/HistoMESA-GUI

Repository files navigation

OncoLlama Test for histo trials

GUI for early user testing of OncoLlama V3. Uses dearpygui.

Python Requirements

  • Python 3.10+ is required (the openpipe dependency uses 3.10+ type syntax).
  • Recommended: Python 3.11.

Install Python 3.11 via pyenv (user-level, no Homebrew)

  1. Install pyenv:
    git clone https://github.com/pyenv/pyenv.git ~/.pyenv
  2. Add to your shell (zsh). Append to ~/.zshrc:
    export PYENV_ROOT="$HOME/.pyenv"
    export PATH="$PYENV_ROOT/bin:$PATH"
    eval "$(pyenv init -)"
  3. Reload shell and install build tools:
    source ~/.zshrc
    xcode-select --install
  4. Install and set Python 3.11 for this project:
    pyenv install 3.11.8
    pyenv local 3.11.8 #run this in the project directory
  5. Create a virtual environment with the correct Python:
    python -m venv .venv
    source .venv/bin/activate
    python --version  # should show 3.11.8

Instructions

  1. Obtain API key and add to .env file
  2. Run pip install -r requirements.txt in a virtual environment.
    • This will also install the oncollamaschemav3 from JTpath/Pathollamaschemav3 package
  3. Run python main.py
  4. If the API returns an error, please retry

Reporting Bad Outputs

The /examples folder contains samples for testing, designed to challenge the model using ambiguous or poor data quality examples

Please report any bad outputs or edge cases to the project team

Note: The configured endpoints in endpoints.yaml are OncoLlama models; the schema/prompt is histopathology-focused, so occasional schema mismatch/validation failures are expected.

About

Fork of londonAI GUI adapting oncollama to histo

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •  

Languages