jlov7's picture
feat: Multi-tool selection and robustness testing
6639f75
raw
history blame contribute delete
480 Bytes
backlog:
- ENV: Set up the development environment with necessary libraries.
- GEN_PAIRS: Generate preference-pair data for training.
- TRAIN: Fine-tune the model using LoFT.
- SCHEMA5: Design 5 unseen schemas for testing.
- TEST: Implement schema_tester.py.
- DEMO: Create demo notebook.
- README: Write README.md with explanations and GIF demo.
sprints:
sprint1:
- ENV
- GEN_PAIRS
- TRAIN
sprint2:
- SCHEMA5
- TEST
- DEMO
- README