Timothyxxx
f9594e476e
Add Support for QWEN models from API (QWEN-max, etc.); Improve on the robustness of getting observation
2024-05-20 00:47:43 +08:00
Timothyxxx
e1e44d77c5
Improve the logic in auto-installation
2024-05-16 18:47:37 +08:00
Timothyxxx
cfdf296913
Fix the reset error when no config inputted; Improve on the robustness of auto installation of vms
2024-05-11 05:36:59 +08:00
Timothyxxx
a59d1f92aa
Improve the logic in auto-installation; Add auto-remove on the failed files
2024-05-10 20:17:40 +08:00
Timothyxxx
e2741c64b0
Improve the logic in auto-installation; Add auto-remove on the failed files
2024-05-10 16:41:20 +08:00
Timothyxxx
bead06afc8
Improve the logic in auto-installation; Add auto-remove on the failed files
2024-05-10 16:24:07 +08:00
zdy023
c1a719d141
ver May1st
...
a temporary workround for Mac hosts.
2024-05-01 21:40:35 +08:00
Timothyxxx
aebbac7285
Increase the waiting time for vm setup
2024-04-25 15:53:12 +08:00
Timothyxxx
b3acf21333
Fix https://github.com/xlang-ai/OSWorld/issues/27
2024-04-23 14:04:59 +08:00
Timothyxxx
0b3e7dca24
Add support for automatic VM download and configuration, enable auto-scaling management; move metadata retrieval out of the init function to speed up environment initialization.
2024-04-21 19:51:15 +08:00
Timothyxxx
9c75df5dce
Clean code; Refactor environment to pass screenshot content instead of path
2024-04-13 23:34:01 +08:00
Timothyxxx
172123ab2c
Support downsampling; Fix bugs in windows a11y tree; Add a11y_tree trim
2024-03-25 18:02:48 +08:00
Timothyxxx
d1e2b12b41
Fix GIMP bug; Speedup the environment, when there is not a11y tree needed, we can do no controller.get
2024-03-20 22:22:59 +08:00
Timothyxxx
8e760fd450
Disable wandb temporarily, speedup the environment step speed by remove useless a11y tree re-get and terminal output
2024-03-19 08:57:05 +08:00
David Chang
2b9772174e
ver Mar15th
...
fixed bugs about infeasible task evaluation
2024-03-15 12:25:41 +08:00
Timothyxxx
35ed7cec89
Update todos
2024-03-14 22:36:33 +08:00
Timothyxxx
44ff027801
Refactor experiments and agent implementation
2024-03-14 22:32:49 +08:00
Timothyxxx
c2aa009ed8
Update server script, baseline and running script
2024-03-13 15:04:19 +08:00
Timothyxxx
81863b26dd
Improve on eval script on web browsing tasks; Add one setup example
2024-02-23 11:57:50 +08:00
Timothyxxx
938a3bb918
Fix get IP error
2024-02-22 20:39:07 +08:00
Timothyxxx
e1cf8da4e0
Fix the infeasible examples support
2024-02-21 21:22:12 +08:00
Timothyxxx
9373773169
add nogui parameter to environment
2024-02-20 17:59:31 +08:00
Timothyxxx
3f59ff46dc
Add infeasible support
2024-02-14 11:59:50 +08:00
Timothyxxx
59e2417a08
Add Mistral, Qwen, Gemini support; Fix minor bugs
2024-02-01 16:55:38 +08:00
rhythmcao
fc15a33b70
finish multi-app examples
2024-02-01 00:53:31 +08:00
Timothyxxx
d65b6994d3
Fix minor bugs of multiple apps examples
2024-01-31 19:40:41 +08:00
David Chang
5a486b6b37
ver Jan27th
...
debugged at+screenshot implementation, no issues found
fixed a little bugs
2024-01-27 23:10:48 +08:00
Timothyxxx
b9ae4174b1
Fix OS examples annotated by Yitao
2024-01-25 19:57:32 +08:00
Timothyxxx
5dea912d01
Finish Chrome v2 loading
2024-01-24 23:05:28 +08:00
Timothyxxx
bdd21d06ca
Fix minor bugs
2024-01-19 20:34:11 +08:00
rhythmcao
91824f754c
1. extend evaluator to list (compatible with single evaluator) 2. fix a variable name error in metrics/general.py
2024-01-18 14:12:54 +08:00
Timothyxxx
b60eb2a933
VM resolution adjust support
2024-01-18 01:43:57 +08:00
Timothyxxx
8efa692951
Add raw accessibility-tree based prompting method (but the tokens are too large); Minor fix some small bugs
2024-01-16 11:58:23 +08:00
Timothyxxx
493b719821
Add gemini agent implementation; Add missed requirements; Minor fix some small bugs
2024-01-15 21:58:33 +08:00
Timothyxxx
f153a4c253
Add 'WAIT', 'FAIL', 'DONE' to the action space; Debug basic prompting-based GPT-4 and Gemini agents; Initialize experiments script;
2024-01-14 23:36:19 +08:00
Timothyxxx
d52b692ee5
Finish loading the vscode examples v1; Improve on the infra: Add accessibility tree into the observation; Add activate window function, etc
2024-01-14 18:30:49 +08:00
Timothyxxx
bc88ee0c41
Minor fix of the logic of vm ip get
2024-01-12 21:18:59 +08:00
rhythmcao
d4116458ff
1. fix quote and \ characters in execute_command ; 2. add terminal output text as extra observation ; 3. move get_vm_*() to reset()
2024-01-12 18:09:05 +08:00
Timothyxxx
5a93a32958
Update on Chrome examples; Refactor on logic of controlling
2024-01-12 17:24:47 +08:00
Timothyxxx
820579a5a2
Make up missing getters and metrics; Update VLC scripts; Start to work on Chrome, update examples instructions
2024-01-11 21:27:40 +08:00
Timothyxxx
287876affc
Merge remote-tracking branch 'origin/main'
...
# Conflicts:
# desktop_env/evaluators/getters/__init__.py
# desktop_env/evaluators/metrics/__init__.py
# requirements.txt
2024-01-10 23:20:49 +08:00
Timothyxxx
49ece15ac3
VLC v1 finished, improve on instructions, improve on infra
2024-01-10 23:18:30 +08:00
David Chang
cf5d480f44
ver Jan10th
...
new Thunderbird task config
2024-01-10 17:36:59 +08:00
Timothyxxx
fa84b20ea5
VLC updates, and some infra bugs fix
2024-01-09 09:30:11 +08:00
David Chang
26b7d9010d
Merge branch 'zdy'
2024-01-05 15:55:41 +08:00
David Chang
eeb8a120d6
ver Jan5th
...
debugged
2024-01-05 15:20:47 +08:00
David Chang
5fedf5b891
ver Jan4th
...
updated interfaces for thunderbird evaluation, not tested
2024-01-04 22:41:57 +08:00
Timothyxxx
ab71ebb2ba
Initialize VLC getters and metrics, fix some bugs in infra logic, needs to be refactored later on
2024-01-04 17:05:17 +08:00
Timothyxxx
03e99a68fb
Loading libreoffice writer examples and find few problems, will do another round tomorrow for the rest
2024-01-02 17:50:05 +08:00
David Chang
4e5920264a
ver Dec27thv2
...
updated a task config
updated documents
fixed the options feature of evaluator
updated with new properties of charts
current load_charts should be ok, I think
2023-12-27 17:51:41 +08:00
