Files
sci-gui-agent-benchmark/run_human_examine.sh
yuanmengqi ea51f5264a fix chrome
2025-06-30 08:07:24 +00:00

9 lines
324 B
Bash

python manual_examine.py \
--headless \
--observation_type screenshot \
--result_dir ./results_human_examine_chrome_fix_1 \
--test_all_meta_path evaluation_examples/test_fix_chrome.json \
--region us-east-1 \
--domain chrome \
--example_id 030eeff7-b492-4218-b312-701ec99ee0cc \
--max_steps 3