Shannon Sands
84e4654795
Fix trailing whitespace and formatting issues in quantum environment documentation - Remove trailing whitespace from code blocks and documentation - Fix end-of-file formatting in README.md - Ensure all pre-commit checks pass for workflow compatibility
2025-05-26 14:14:36 +10:00
Shannon Sands
b845c635d4
linted, moved to community folder
2025-05-26 14:10:26 +10:00
Shannon Sands
20c6e9d8d7
Merge remote-tracking branch 'jeannemtl/hack/env_quant' into merge-jeannemtl-contributions
2025-05-26 13:52:36 +10:00
Shannon Sands
aff033443f
linting
2025-05-26 13:42:14 +10:00
Shannon Sands
7cfd3af149
Integrate Caput Mundi poker environment from yoniebans - Add Six-Seat No-Limit Hold'em poker training environment - Features expert hand history training with dual reward system - Includes action matching and bet sizing evaluation components - Supports multi-stage game analysis (preflop/flop/turn/river) - Integrates with HuggingFace datasets and WandB monitoring - Comprehensive documentation added to community README ( #17 ) - All code quality checks passing (black, isort, flake8) Environment moved from hack0/poker to environments/community/poker_holdem/ Resolves PR #84 from yoniebans/atropos
2025-05-26 13:38:49 +10:00
Shannon Sands
04c06c3e20
Merge remote-tracking branch 'yoniebans/main' into merge-yoniebans-contributions
2025-05-26 13:33:36 +10:00
Shannon Sands
0f61c9dbde
moved to community folder
2025-05-26 13:27:43 +10:00
Shannon Sands
a17dbdfedc
Merge remote-tracking branch 'metonym/deepsacrifice' into merge-metonym-contributions
2025-05-26 13:22:22 +10:00
Shannon Sands
3707ac939f
linting
2025-05-26 13:03:23 +10:00
Shannon Sands
bc1f85619f
Fixed linting issues
2025-05-26 12:59:55 +10:00
Shannon Sands
f17c07c823
Merge remote-tracking branch 'justin5764/LeanRL' into merge-justin5764-contributions
2025-05-26 12:47:52 +10:00
Shannon Sands
5551580170
resolved conflicts
2025-05-26 12:42:37 +10:00
Shannon Sands
5d22d360e2
Add Solitaire Winning Probability Environment - Mathematical probability analysis environment for training LLMs - Combines theoretical formula derivation with Monte Carlo simulation - Supports various solitaire-style card games - Includes sophisticated reward system with relative error calculation - All API keys removed for security - Comprehensive documentation added to community README - Author: davidedipeppe, PR: #88
2025-05-26 12:36:24 +10:00
Shannon Sands
d789128f20
Fix final code quality issues in Conversational Style DPO environment
2025-05-26 10:48:11 +10:00
Shannon Sands
441fd1036d
Merge Karthik-Ragunath conversational style DPO environment contribution
2025-05-26 10:25:08 +10:00
Shannon Sands
c2c4928882
Fix final line length violations in Pokemon Showdown environment
2025-05-26 10:15:32 +10:00
Shannon Sands
0038a710d0
merging
2025-05-26 09:58:38 +10:00
Shannon Sands
c360ee20e7
linting
2025-05-26 09:39:51 +10:00
Shannon Sands
65108d12b2
Linting done
2025-05-26 09:28:23 +10:00
Shannon Sands
a58562447f
Merge branch 'joshuajerin-selcube' into merge-joshuajerin-contributions
2025-05-26 09:07:25 +10:00
Shannon Sands
abab64e8dc
full commit
2025-05-26 08:48:33 +10:00
Shannon Sands
129b310593
Integrate JakeBoggs punchline VR-CLI environment - Add Punchline VR-CLI environment for training humor understanding using VR-CLI methodology - Moved from environments/hack0/punchlines to environments/community/punchline_vrcli - Updated community README with comprehensive environment description - Fixed linting issues and formatted code per project standards - Credit: @JakeBoggs
2025-05-26 08:45:45 +10:00
Shannon Sands
c3e2046a20
Merge branch 'JakeBoggs-punchline' into merge-jakeboggs-contributions
2025-05-26 08:33:37 +10:00
Teknium
4b532da35e
Merge pull request #114 from leopardracer/main
...
Improve API Server Documentation and Update UFC Prediction Output Format
2025-05-25 01:34:48 -07:00
Shannon Sands
0c4c3e1e6c
linting
2025-05-24 14:43:24 +10:00
Shannon Sands
47c42bdc72
linting
2025-05-24 14:37:24 +10:00
Shannon Sands
160abf8574
Integrate krishpop's Cat Behavior Communication Environment - Merged cat behavior environment from krishpop:main - Moved cat files from environments/ to environments/community/cat_behavior_env/ - Fixed file paths for cat_behaviors.json and cat_scenarios.json - Removed unused imports and fixed all linting issues - Updated community README with comprehensive cat environment description - Credited author @krishpop with GitHub link
2025-05-24 14:21:58 +10:00
Shannon Sands
f399e3513f
Merge remote-tracking branch 'krishpop/main' into merge-krishpop-contributions
2025-05-24 13:54:43 +10:00
Shannon Sands
95bec5e7a8
Integrate RoshanSanjeev's ExamCraft environment - Merged ExamCraft environment from RoshanSanjeev PR #95 - Moved from environments/hack0/ to environments/community/ - Removed demo_artifacts.tar.gz file to avoid repo clutter - Updated community README with comprehensive ExamCraft description - Fixed all linting issues (flake8, black, isort) - Credited author @RoshanSanjeev with GitHub link
2025-05-24 13:48:52 +10:00
Shannon Sands
455fbd053c
Merge branch 'RoshanSanjeev-examcraft' into merge-roshansanjeev-contributions
2025-05-24 13:39:03 +10:00
Shannon Sands
32cf5e3d42
Integrate joshgarza's accessibility environment - Merged accessibility environment from joshgarza:main - Moved from environments/hack0/ to environments/community/ - Updated community README with detailed description of accessibility auto-fixer - Added note about missing dataset file - Credited author @joshgarza with GitHub link
2025-05-24 13:31:50 +10:00
Shannon Sands
30ddc8a36d
Merge remote-tracking branch 'joshgarza/main' into merge-joshgarza-contributions
2025-05-24 13:29:23 +10:00
teknium1
ae0340bb9f
prevent token explosion issue by reducing max_token to 15k instead of 16k
2025-05-23 18:09:36 -07:00
teknium1
1fa798a69e
Making saving data optional in config, add scores to saved data
2025-05-23 14:11:11 -07:00
teknium1
a20886d720
fix many many things jules didnt do right
2025-05-23 12:50:38 -07:00
leopardracer
c15149c23b
Update ufc_image_env.py
2025-05-23 19:43:04 +03:00
Shannon Sands
8aff7cb3c5
Fix import sorting issues in environment files - Move wandb imports to third-party section in philosophical_rlaif_env.py, math_server.py, math_server_zero.py, and tool_calling_server.py - Add missing spaces after commas in ufc_server.py stats.get() calls - Manual fixes to resolve CI linting failures
2025-05-23 15:58:16 +10:00
Shannon Sands
7de20bddfe
merged changes
2025-05-23 15:48:14 +10:00
Shannon Sands
6e55cbc448
Add edmundman's UFC prediction environment with sample dataset - Moved UFC environment from hack0/ to community/ufc_prediction_env/ - Fixed all linting issues: unused imports, long lines, unused variables - Trimmed large_dataset.csv to 799 records (459KB) to meet repository limits - Added comprehensive documentation to community README - Environment features both text-based and image-based fight prediction - Generates entertaining TTS-ready commentary for voice synthesis - Includes web scraping tools and Flask UI interface
2025-05-23 15:47:12 +10:00
Shannon Sands
606b917042
Merge edmundman's UFC_FIGHT_PREDICTOR contribution
2025-05-23 15:33:02 +10:00
Fabrizio (Misto) Milo
490caf6854
change directory
2025-05-22 13:03:37 -07:00
teknium1
46d33bf0b2
manually implement readme update due to
2025-05-21 23:09:45 -07:00
shannonsands
ba4ee8d68b
Merge pull request #63 from NousResearch/toolcalling-server-process-fix
...
Toolcalling server process fix
2025-05-21 21:56:11 -07:00
based-tachikoma
b01023ad3a
additional fixes to alphafold2_multimer and tool_executor
2025-05-21 21:28:30 -07:00
based-tachikoma
6783a077cc
update comment in protein_env.py
2025-05-21 21:27:42 -07:00
based-tachikoma
02585947e4
refactor file saving from alphafold2_multimer to tool_executor
2025-05-21 21:23:00 -07:00
Shannon Sands
a614fa7e67
updated llms.txt, contribution guide and added community folder with README
2025-05-22 11:53:11 +10:00
google-labs-jules[bot]
276a845dd7
feat: Implement SWE-RL Environment with Full Refinements
...
I've implemented the SWERLEnv in environments/swe_rl_env.py, based on the
SWE-RL paper (arXiv:2502.18449). This version incorporates extensive
refinements based on your feedback.
Key features implemented in environments/swe_rl_env.py:
- Core environment structure (setup, trajectory collection, scoring, evaluation).
- "Thinking" step: LLM is prompted for reasoning within <think> </think> tags
before generating a patch. Includes strict parsing for these tags.
- Dynamic prompt construction using `tokenizer.apply_chat_template` with
NousResearch/DeepHermes-3-Llama-3-8B-Preview as the default model.
- Hugging Face dataset integration: Loads data from HF Hub with configurable
dataset name, splits, and column mappings.
- Reward mechanism: Based on thinking tag correctness, patch format
(SEARCH/REPLACE), and similarity to the oracle patch.
- Comprehensive WandB logging for training/evaluation metrics.
NOTE: I made multiple attempts to update 'environments/README.md'
with documentation for this new environment. While I
reported success in some turns, this was not consistently verifiable
and may not have been correctly applied. The README.md file may
require manual verification and updating for the SWERLEnv.
2025-05-22 01:28:00 +00:00
Andrew
5d34ea821d
removed html from data folder
2025-05-21 18:10:10 -07:00
Andrew
6316cf31a5
chore: remove .zip and .html files per review feedback
2025-05-21 18:08:22 -07:00
