ryanhoangt/OpenHands-evaluation

add missing report file

add CoAct v1.0 trajectories

Add CoAct v1.0 trajectory (#10)

add collapsable

update instruction

remove all the with hint result

cleanup metrics and fix repo

cleanup metrics and fix repo

add llama-3.1 result

rename file

rename OpenDevin to OpenHands

bump streamlit ver

fix typo

update app file

fix visualizer with latest streamlit feature

add 2nd run

add gpt-4o-mini result

Revert "add result from gpt-4o-mini"

add result from gpt-4o-mini

update the last missing instance

update result from pr2489

remove keys

revoke keys

add gpqa result

update v1.8 perf

add result for v1.8 no-hint gpt4o

fix model_name in updated metadat

add v1.8 result

update results using new ver of swebench

set n error/stuck/cost to 0 for CodeAct exp run below v1.5

by default not showing with hint result

add claude-3.5 result

support loading report with new format

update gitignore

update old result w/ swe-bench latest harness;

improved patch apply

improved patch apply

add report field

Add CodeAct 1.6 no hint

fix visualizer

feat: add gpqa results (#8)

fix visualizer to only display eval_report when it exists

add result for codeact 1.6

only show swe bench on visualizer

change test_result to bool

fix fine-grained report; support visualization while running

add gpt-4-1106 results for codeact swe

Merge commit 'edc3858a6ea5d0c7317b630024203af60e146b52'

update all swebench lite

Update outputs/miniwob/README.md