Fix health check: use Python instead of curl

Fix HuggingFace Space deployment: install curl for health checks and increase startup timeout

Fix healthcheck to use correct port with PORT environment variable

Fix root path 404: add redirect from / to /ui/

Fix HuggingFace Space deployment: use same-origin WebSocket and dynamic port

Fix WebSocket protocol for HTTPS deployment: use wss:// for secure connections

Rewrite README with philosophical, accessible prose for broader audience

formated readme

Clean up submission links table formatting

Update README with HF Space URL and submission links

Remove VS cache files, add to gitignore

Add blog post and HF submission checklist for hackathon

Model name update

Updated readme

Document Colab 100-step training summary in README

Readme cleanup

Add 100-step Colab GRPO training results

Fix eval decode prompt length slicing

Fix GRPO reward mapping and evaluation generation

Include prompt column for GRPOTrainer

Use finite datasets.Dataset for GRPOTrainer compatibility

Yield tensors from GRPO dataset generator

Fix GRPO dataloader batching on Kaggle

Fix GRPO sample id mapping and Kaggle training setup

Add sample training results and judge report for hackathon submission

Fix division by zero in final avg improvement calculation

Fix category enum: use lowercase for ProbeAction categories

Fix eval_report.py: use step() return value and extract reward from observation

Remove unsupported data_collator argument from GRPOTrainer

Add data collator to GRPOTrainer for on-the-fly prompt tokenization

Revert to raw prompt format—let GRPOTrainer handle tokenization

Convert tokenized outputs to torch tensors for proper batching

Tokenize prompts in dataset generator for GRPOTrainer compatibility

Fix prompt format: convert from chat list to string for GRPOTrainer compatibility

Fix dataset format: only yield 'prompt' field to avoid tensor concatenation errors

Remove tokenizer argument from GRPOTrainer—not supported in installed TRL version

Fix path bootstrap in train_grpo.py: use parent.parent to reach project root

Fix GRPO batch size config for Kaggle P100: batch_size=4, grad_accum=2 (global=8, divisible by num_generations=2)

Add eval_report.py for before/after training comparison

Add pre-training baseline results and graphs

Fix JSON parsing and environment bugs

BlogPost for Huggeing later use

Updated readme

UI Integration

Added Meaning ful comments

Code improvenets

Updated readme file

refactor: remove legacy architecture, promote clean structure to repo root

Worked on folder structure

Judge matching chnages with name changes