reflexion-human-eval/run_testacc.sh
2023-04-11 20:15:57 -04:00

11 lines
250 B
Bash
Executable File

python main.py \
--run_name "testacc_humaneval_py3" \
--root_dir "root" \
--dataset_path ./benchmarks/humaneval-py.jsonl.gz \
--strategy "test-acc" \
--language "py" \
--model "gpt-4" \
--pass_at_k "1" \
--max_iters "1" \
--verbose