autogen

Commit Graph

Author	SHA1	Message	Date
Adam Fourney	5acf613fd4	Fix some issues with the WebArena Orchestrator	2 years ago
Adam Fourney	b455ab9c3e	Use the cheaper and faster TwoAgent protocol to initially login and set up the main task.	2 years ago
Adam Fourney	3824685272	Align Orchestrator settings to match TwoAgents	2 years ago
afourney	b068528bc9	Merge pull request #2500 from microsoft/ct_webarena-dockerfix Fix Dockerfile + Add response_format to Template + Fix requirement.txt	2 years ago
afourney	5b47f0196b	Merge pull request #2493 from microsoft/webarena_split Webarena add Test Validation split	2 years ago
gagb	363ad107fe	removed unnecessary platform specific dependencies	2 years ago
gagb	998e918e02	Update orch template to use response format var	2 years ago
Adam Fourney	b2463077ab	Added comments to explain the test/validation split logic.	2 years ago
Eduardo Salinas	035659f944	rename args	2 years ago
Eduardo Salinas	6520505239	code fmt	2 years ago
Eduardo Salinas	bd2b10ea88	Update Dockerfile	2 years ago
Eduardo Salinas	bb30c7babe	Merge branch 'ct_webarena' into add-firefox	2 years ago
Eduardo Salinas	32e59b1e66	add firefox to scenario.py and other tweaks	2 years ago
Adam Fourney	34fa8b0ade	Split WebArena into test and validation sets, based on template identifier.	2 years ago
Adam Fourney	2da2b586d1	Added punkt dependency to Dockerfile.	2 years ago
afourney	18eba94d14	Merge pull request #2468 from microsoft/gagb-ct_webarena Add Orchestrator to WebArena	2 years ago
Gagan Bansal	f217348200	Remove unused file	2 years ago
Gagan Bansal	55b9429be3	Remove unused file	2 years ago
Gagan Bansal	e15ba0ace3	Add initial autogenbench template for Orchestrator	2 years ago
Gagan Bansal	567ccc1025	Run pre-commit on all files	2 years ago
afourney	9ad2b7bfcc	Merge pull request #2399 from microsoft/gagb-ct_webarena Improve documentation of MMWebsurfer	2 years ago
Eduardo Salinas	9f324e9011	rename event	2 years ago
Eduardo Salinas	19df31d540	fixup	2 years ago
Eduardo Salinas	86ea88d6e6	fixup	2 years ago
Eduardo Salinas	f5f96cbb63	add to scenario.py	2 years ago
Eduardo Salinas	628e47c0bc	add logging to multimodal_web_surfer	2 years ago
Gagan Bansal	0646a6257e	Run precommit	2 years ago
gagb	4a083b81e4	Ignore results dir	2 years ago
Adam Fourney	0c7dffaf4d	Updated Dockerfile for webarena	2 years ago
Adam Fourney	d579a7d51c	Added tabulation for WebArena	2 years ago
Adam Fourney	891fc9b62b	Addressed some issues with scoring.	2 years ago
Adam Fourney	6804bc2ff5	Added eval code to webarena	2 years ago
Adam Fourney	1a08fd7b24	Added prompting to indicate to agents that they are on Postmill.	2 years ago
Adam Fourney	e7944ae75d	run_cmd.py	2 years ago
Adam Fourney	d9dd280e7b	Added the ability to scroll controls, and fixed a bug with option names.	2 years ago
Adam Fourney	285e47c511	Added sample ENV file, and cleaned up scenario.py	2 years ago
afourney	061a857b3d	AutoGenBench: Handle Ctrl-C more gracefully. (#2174 ) * Prints the version of AutoGenBench from the command line, closing i1458 * Added autogenbench version to timestamp.txt * Attempting to fix formatting. * Add a gitignore for autogenbench * Generalize to read all template dirs from Templates * AutoGenBench logs telemetry when available. * Remove spaces if present from template names. * Bump version. * Fixed formatting. * Allow native warning to be skipped. Mount autogen repo in Docker if it can be found (experimental). * Native execution now occurs in a venv. * Bump version. * Fixed a prompt escaping bug evident in GAIA task '6f37996b-2ac7-44b0-8e68-6d28256631b4' * Updated all scenarios to use template discovery. * Update with main version of runtime_logging. * Better handling of Ctrl-C and cleanup of unused containers. * Even stronger hinting that containers should be removed. --------- Co-authored-by: gagb <gagb@users.noreply.github.com>	2 years ago
Adam Fourney	a3d26e918f	Merge branch 'autogenbench' into ct_webarena	2 years ago
Adam Fourney	618533a0ac	Added debug dir for websurfer output.	2 years ago
Adam Fourney	92282ce186	Better handling of Ctrl-C and cleanup of unused containers.	2 years ago
Adam Fourney	ab04185f63	Merge branch 'multimodal_web_surfer' into ct_webarena	2 years ago
Adam Fourney	765fcd3281	Initial work on onboard WebArena to AutoGenBench	2 years ago
olgavrou	af9b300be3	add webarena in samples (#2114 ) * add webarena in samples/tools * Update samples/tools/webarena/README.md Co-authored-by: gagb <gagb@users.noreply.github.com> * Update samples/tools/webarena/README.md Co-authored-by: gagb <gagb@users.noreply.github.com> * Update samples/tools/webarena/README.md Co-authored-by: gagb <gagb@users.noreply.github.com> * update installation instructions * black formatting * Update README.md --------- Co-authored-by: gagb <gagb@users.noreply.github.com> Co-authored-by: Eric Zhu <ekzhu@users.noreply.github.com>	2 years ago
Davor Runje	b1839c3845	Update pre-commit (#2067 ) * update pre-commit * update pre-commit.ci * lint fix	2 years ago
afourney	3a5dd361b9	Bump autogenbench version. (#2027 )	2 years ago
Eduardo Salinas	6dbae0a88b	fix: [autogenbench] writing to stdout encoding error in win-os (#2002 )	2 years ago
Eduardo Salinas	a814ba54de	fix: [autogenbench] windows fails unless we specify encoding (#1957 )	2 years ago
olgavrou	ce71d85e77	Ability to fine tune custom model on conversable agents (#1787 ) * uAbility to update_model on conversable agents * formatting * formatting * move code from conversable agent into samples/tools and add testing and README * forgot install step * fix * leave core lib unchanged and move everything to samples/tools * remove skip openai --------- Co-authored-by: Eric Zhu <ekzhu@users.noreply.github.com>	2 years ago
Yiran Wu	2503000c22	update (#1891 )	2 years ago
afourney	085bf6cf3d	Version 0.0.2 of Autogenbench (#1548 ) * Prints the version of AutoGenBench from the command line, closing i1458 * Added autogenbench version to timestamp.txt * Attempting to fix formatting. * Add a gitignore for autogenbench * Generalize to read all template dirs from Templates * AutoGenBench logs telemetry when available. * Remove spaces if present from template names. * Bump version. * Fixed formatting. * Allow native warning to be skipped. Mount autogen repo in Docker if it can be found (experimental). * Native execution now occurs in a venv. * Bump version. * Fixed a prompt escaping bug evident in GAIA task '6f37996b-2ac7-44b0-8e68-6d28256631b4' * Updated all scenarios to use template discovery. * Update with main version of runtime_logging. --------- Co-authored-by: gagb <gagb@users.noreply.github.com>	2 years ago

1 2

65 Commits (59ab0f2bb2cb71d6a67e759549bb9367daee627a)