Releases: ServiceNow/AgentLab
Releases · ServiceNow/AgentLab
v0.4.0
What's Changed
- chore: update init.py by @eltociear in #181
- parallel study evaluation by @recursix in #180
- Refactor HuggingFace model initialization to include base model name … by @jardinetsouffleton in #190
- limiting to python 3.11 and above by @ThibaultLSDC in #194
- Implement parallel processing for studies using ProcessPoolExecutor a… by @recursix in #195
- Update README.md, fix typo by @ollmer in #196
- small API change - passing exp_root to study.run() by @optimass in #200
- added
study.shuffle_exps()
feature by @optimass in #202 - trying to fix tests by @ThibaultLSDC in #206
- Add new agent configurations for Claude Sonnet 3.5 and vision models by @jardinetsouffleton in #213
New Contributors
- @eltociear made their first contribution in #181
- @optimass made their first contribution in #200
Full Changelog: v0.3.2...v0.4.0
v0.3.2
What's Changed
- displaying exp names in ray dashboard by @ThibaultLSDC in #123
- Fixing goal not being used in ui_assistant mode by @ThibaultLSDC in #124
- Fixing discussion object when adding images w/o detail by @ThibaultLSDC in #128
- Adding descriptive prompts for screenshot/som by @ThibaultLSDC in #129
- Study to multi eval by @recursix in #126
- Update README.md by @recursix in #158
- Update README.md by @recursix in #159
- Enhance README with examples for loading experiment results by @recursix in #160
- Warning notice + link to BrowserGym by @gasse in #164
- 405b results on workarena L2 by @ThibaultLSDC in #163
- Ab res by @ThibaultLSDC in #161
- Add fix for self-hosted HF models by @jardinetsouffleton in #167
- WebArena/VisualWebArena results by @ThibaultLSDC in #168
- Adding suffix to tracker decorator by @ThibaultLSDC in #169
- Multiple output chat and retry function by @ThibaultLSDC in #171
- For webarena agent by @recursix in #172
- fixing pypi workflow dependency by @ThibaultLSDC in #174
- fix: update demo_mode assignment in GenericAgentArgs class by @recursix in #175
- Adapt multiple samples for HF models by @jardinetsouffleton in #173
- automated readthedocs by @ThibaultLSDC in #177
Full Changelog: v0.3.1...v0.3.2
v0.3.2.dev9
version bump
v0.3.2.dev11
last one
v0.3.2.dev10
tmp updates just for a check
v0.3.2.dev7
one last test
v0.3.2.dev6
testing some more stuff
v0.3.2.dev5
version
v0.3.2.dev4
cleaning up
v0.3.2.dev3
version bump