Download code/run_all.sh from shewbham/firstaid-sim: direct link, hf CLI and curl.
- Browser
- Download file 435 Bytes
-
https://huggingface.co/datasets/shewbham/firstaid-sim/resolve/main/code/run_all.sh
- Command line
-
hf download hf://datasets/shewbham/firstaid-sim/code/run_all.sh
-
curl -L -o run_all.sh https://huggingface.co/datasets/shewbham/firstaid-sim/resolve/main/code/run_all.sh
435 Bytes
| # Base eval -> LoRA SFT -> fine-tuned eval, back to back. | |
| set -e | |
| M=CohereLabs/tiny-aya-fire | |
| T=artifacts/scenarios_test6.jsonl | |
| export HF_HUB_OFFLINE=1 PYTORCH_ENABLE_MPS_FALLBACK=1 | |
| .venv/bin/python evaluate.py --name base --model $M --scenarios $T | |
| .venv/bin/python train.py --model $M --out out/lora --minutes ${TRAIN_MIN:-22} | |
| .venv/bin/python evaluate.py --name sft --model $M --adapter out/lora --scenarios $T | |
| echo ALL_DONE | |