Beginner's Guide (Start Here)¶
Welcome! This guide is for developers new to Python ML projects. You'll get a working demo in minutes, no GPU required.
Prerequisites¶
- Python 3.8–3.11
- Git (optional)
1) Set up your environment¶
# inside your project folder
python -m venv .venv
# Windows: .venv\\Scripts\\activate
# macOS/Linux:
source .venv/bin/activate
pip install -r requirements.txt
pip install -r requirements-dev.txt
pip install -e .
pre-commit install
2) Run the beginner demo (no downloads)¶
Option A (Makefile): This creates a sample run folder with the files the Studio viewer expects.
make beginner
Option B (CLI only):
llm-ripper quickstart --open
The command will:
- Generate a run under runs/<timestamp>/ with demo JSON files
- Launch the Studio viewer at http://localhost:8000
If the page shows empty panels or error messages, that's okay — you can still explore the layout and JSON files.
3) Next steps¶
- Try the offline smoke test:
make smoke-offline - Explore CLI help:
python -m llm_ripper.cli --help - Read the Quickstart for full pipeline steps
Troubleshooting¶
- If
mkdocsorruffcommands are missing, install dev deps:pip install -r requirements-dev.txt - If ports are busy, change the Studio port:
make studio PORT=8001
You're set! As you gain confidence, switch from the demo to actual models and data.