Spaces:

Leavin1611
/

logistics-hackathon-env

Sleeping

App Files Files Community

logistics-hackathon-env

Commit History

Apr 26

10:20
docs: add slide deck link to blog post Quick Start section

Jerry

1f8b5c8
10:14
feat: add /slides route to FastAPI so slide deck is viewable on HF Space

Jerry

69d7ed0
10:10
feat: add 9-slide HTML presentation deck and link in README

Jerry

c57da4f
10:05
docs: add prominent model links to README and blog post

Jerry

95b0b90
09:59
docs: update DESIGN.md with final hackathon architecture and results

Jerry

540b7d9
09:57
Upload assets/training_reward_curve.png with huggingface_hub

Leavin1611

Verified signature
7f24ee1
09:45
docs: add training reward curve to blog post

Jerry

81ffb8a
09:42
docs: make evaluation graphs visible by default in README and add them to blog post

Jerry

9805dbf
09:37
Upload assets/eval_logic.png with huggingface_hub

Leavin1611

Verified signature
6f97ab4
09:37
Upload assets/eval_hacking.png with huggingface_hub

Leavin1611

Verified signature
607ef00
09:37
Upload assets/eval_efficiency.png with huggingface_hub

Leavin1611

Verified signature
cc5bae9
09:37
Upload assets/eval_curriculum.png with huggingface_hub

Leavin1611

Verified signature
d45d0e1
09:32
docs: remove video placeholder and add actual trained model link

Jerry

0a4e759
08:45
fix: repair final 2 syntax errors in cells 24 and 43

Jerry

b975d31
08:44
fix: repair all 14 syntax errors across notebook cells

Jerry

7e71b14
08:33
fix: replace all literal backslash-n in import lines with real newlines

Jerry

4735d91
08:28
fix: add self-contained imports to all cells that use torch/requests/np

Jerry

296b0d2
08:26
fix: make eval cells self-contained with explicit imports

Jerry

206eca6
08:18
Upload assets/training_reward_curve.png with huggingface_hub

Leavin1611

Verified signature
5c6f5db
08:09
fix: corrected indentation error in run_model_episode cell

Jerry

874cab4
08:01
docs: added detailed evaluation and ablation study graphs to README gallery

Jerry

5655640
07:49
docs: finalize and clean Colab notebook with markdown structure

Jerry

9bda2dc
05:52
docs: update README and blog with real 0.7683 score, reward curve embed, fixed Colab links

Jerry

910aacf
04:15
feat: add training reward curve plot cell with phase separators and rolling average

Jerry

7e8e6fc
03:15
feat: replace CoT block with advanced non-CoT fine tuning phase

Jerry

7517555
03:04
feat: restructure notebook to skip early phases and start directly from Phase 3 hardened model for advanced fine-tuning

Jerry

74ad613
02:57
feat: boost training hyperparameters (datasets, epochs, generations) for maximum accuracy

Jerry

7b38082
02:55
revert: remove CoT phase entirely to restore original high-performing 3-phase curriculum

Jerry

c358fa5
02:38
fix: remove surrogate pairs from print statements to prevent Colab UnicodeEncodeError

Jerry

2aa6cd5
00:32
docs: add hackathon evaluation criteria checklist to README

Jerry

bd633bd

Apr 25

22:39
fix: remove literal newline and formatting characters from Phase 4 code cells

Jerry

c8d97da
22:26
feat: append Phase 4 to the executed notebook for judging

Jerry

b9d7eb6
22:21
feat: append Advanced CoT Fine-Tuning as Phase 4 for judges

Jerry

df9e8a9
22:17
feat: simplify training loop to single CoT fine-tuning phase

Jerry

8c8b9ff
22:12
feat: fine-tune from previous LoRA, add CoT reasoning, increase dataset size

Jerry

e7cfcaa
22:01
fix: only update background traffic congestion at end of turn

Jerry

d7e4abb
14:46
Fix reward hacking loophole and add Phase 3 Hardening

Jerry

0378774
12:39
Production: 3 reward fns, anti-hack penalties, output inspection, LoRA saving, try/except

Jerry

9ad6b29
11:41
Feature: Add Curriculum Learning workflow (Phase 1 Easy -> Phase 2 Medium)

Jerry

04e6d1e
11:35
Final Fix: Switch to Analytic Reward Function to guarantee convergence

Jerry

74980cb
09:30
Fix: Correct GRPO - sequence-level rewards, penalize action spamming

Jerry

9e20b7b
09:03
Final: Probe-then-Train GRPO with real HF Space reward signal

Jerry

eafe0b1
08:49
Fix: Context-aware GRPO rewards, eliminate training collapse (-100% bug)

Jerry

109ec41
08:09
Final: Stable baseline, env-grounded GRPO, comparison graph export

Jerry

adfe7d2
07:56
Fix: Ground GRPO rewards in live environment API calls (eliminate reward misalignment)

Jerry

42bea6d

Apr 24

19:52
Final: Grand Finale Training with Chain-of-Thought reasoning

Jerry

62e1682
19:20
Update: Turbo Mode Training for higher reward delta

Jerry

4686178
19:01
Fix: Unwrap observation in Colab notebook API calls

Jerry

f1ac142
18:10
feat: Anti-hack penalty, storytelling README, HF blog post — judges criteria compliance

Jerry

aed246b

Apr 23

22:43
Fix: Unwrap observation from API response in dashboard; add local setup/start scripts

Jerry

048d900

Commit History

Apr 26

docs: add slide deck link to blog post Quick Start section

feat: add /slides route to FastAPI so slide deck is viewable on HF Space

feat: add 9-slide HTML presentation deck and link in README

docs: add prominent model links to README and blog post

docs: update DESIGN.md with final hackathon architecture and results

Upload assets/training_reward_curve.png with huggingface_hub

docs: add training reward curve to blog post

docs: make evaluation graphs visible by default in README and add them to blog post

Upload assets/eval_logic.png with huggingface_hub

Upload assets/eval_hacking.png with huggingface_hub

Upload assets/eval_efficiency.png with huggingface_hub

Upload assets/eval_curriculum.png with huggingface_hub

docs: remove video placeholder and add actual trained model link

fix: repair final 2 syntax errors in cells 24 and 43

fix: repair all 14 syntax errors across notebook cells

fix: replace all literal backslash-n in import lines with real newlines

fix: add self-contained imports to all cells that use torch/requests/np

fix: make eval cells self-contained with explicit imports

Upload assets/training_reward_curve.png with huggingface_hub

fix: corrected indentation error in run_model_episode cell

docs: added detailed evaluation and ablation study graphs to README gallery

docs: finalize and clean Colab notebook with markdown structure

docs: update README and blog with real 0.7683 score, reward curve embed, fixed Colab links

feat: add training reward curve plot cell with phase separators and rolling average

feat: replace CoT block with advanced non-CoT fine tuning phase

feat: restructure notebook to skip early phases and start directly from Phase 3 hardened model for advanced fine-tuning

feat: boost training hyperparameters (datasets, epochs, generations) for maximum accuracy

revert: remove CoT phase entirely to restore original high-performing 3-phase curriculum

fix: remove surrogate pairs from print statements to prevent Colab UnicodeEncodeError

docs: add hackathon evaluation criteria checklist to README

Apr 25

fix: remove literal newline and formatting characters from Phase 4 code cells

feat: append Phase 4 to the executed notebook for judging

feat: append Advanced CoT Fine-Tuning as Phase 4 for judges

feat: simplify training loop to single CoT fine-tuning phase

feat: fine-tune from previous LoRA, add CoT reasoning, increase dataset size

fix: only update background traffic congestion at end of turn

Fix reward hacking loophole and add Phase 3 Hardening

Production: 3 reward fns, anti-hack penalties, output inspection, LoRA saving, try/except

Feature: Add Curriculum Learning workflow (Phase 1 Easy -> Phase 2 Medium)

Final Fix: Switch to Analytic Reward Function to guarantee convergence

Fix: Correct GRPO - sequence-level rewards, penalize action spamming

Final: Probe-then-Train GRPO with real HF Space reward signal

Fix: Context-aware GRPO rewards, eliminate training collapse (-100% bug)

Final: Stable baseline, env-grounded GRPO, comparison graph export

Fix: Ground GRPO rewards in live environment API calls (eliminate reward misalignment)

Apr 24

Final: Grand Finale Training with Chain-of-Thought reasoning

Update: Turbo Mode Training for higher reward delta

Fix: Unwrap observation in Colab notebook API calls

feat: Anti-hack penalty, storytelling README, HF blog post — judges criteria compliance

Apr 23

Fix: Unwrap observation from API response in dashboard; add local setup/start scripts