Commit History

Sync with github: Training results and advanced RLVR environment
0b07253
Running
verified

musharraf7 commited on

Sync with github: Training results and advanced RLVR environment
cf86f90
verified

musharraf7 commited on

Sync with github: Training results and advanced RLVR environment
3381f43
verified

musharraf7 commited on

fix: redirect root to /demo/
38cc60a
verified

musharraf7 commited on

fix: remove Gradio 6 css warning
9e1258c
verified

musharraf7 commited on

ui: redirect root directly to /demo
65dfc27
verified

musharraf7 commited on

ui: rebuild ESCTR frontend with direct app landing and blog tab
1dcdbe0
verified

musharraf7 commited on

ui: refresh Gradio interface with cleaner editorial design
fdb62ab
verified

musharraf7 commited on

fix: remove uv.lock from gitignore
7249f93
verified

musharraf7 commited on

fix: include uv.lock (required by openenv validate)
da58d96
verified

musharraf7 commited on

polish: Blog.md final storytelling + absolute links
6ba8d64
verified

musharraf7 commited on

docs: add Phase 4 (1.7B HF Jobs) to Blog.md
68287c5
verified

musharraf7 commited on

docs: add 1.7B run metrics to README
a8b2fa7
verified

musharraf7 commited on

docs: add all 3 training scripts to README
3662e38
verified

musharraf7 commited on

fix: make get_reward/is_done private to hide from TRL tool schema
25b3086
verified

musharraf7 commited on

fix: use ESCTRToolEnv wrapper + correct reward signature from train_4b.py
69e37b3
verified

musharraf7 commited on

fix: add jmespath dep
dbefdec
verified

musharraf7 commited on

fix: restore LoraConfig constructor, remove model_init_kwargs
7fb792f
verified

musharraf7 commited on

fix: remove max_prompt_length from GRPOConfig
58e71db
verified

musharraf7 commited on

fix: remove dead Action import
285ca7e
verified

musharraf7 commited on

fix: torch>=2.6 for TRL FSDPModule compat
883c1d7
verified

musharraf7 commited on

fix: correct openenv version and local install path
143a921
verified

musharraf7 commited on

feat: add HF Jobs training script
501bf5f
verified

musharraf7 commited on

docs: fix all submission links to absolute URLs
cec6b17
verified

musharraf7 commited on

docs: sync Blog.md with user edits
671b402
verified

musharraf7 commited on

docs: remove dead 4B weights links from Blog.md
68cdd00
verified

musharraf7 commited on

Delete egg-info
7413d1e
verified

musharraf7 commited on

Delete uv.lock
c8dee49
verified

musharraf7 commited on

Delete blog_post.md
cc2642d
verified

musharraf7 commited on

Delete hf_upload.py
162d0a0
verified

musharraf7 commited on

Delete SUBMISSION_CHECKLIST.md
de1c337
verified

musharraf7 commited on

Delete PLAN.md
56aa310
verified

musharraf7 commited on

Delete Academic framing: what to cite and how to position ESCTR.txt
ec8433e
verified

musharraf7 commited on

Delete Untitled
b6b037f
verified

musharraf7 commited on

Upload folder using huggingface_hub
08a3b81
verified

musharraf7 commited on

Upload folder using huggingface_hub
8a4babc
verified

musharraf7 commited on

Sync with github: Training results and advanced RLVR environment
af7c75f
verified

musharraf7 commited on

Sync with github: Training results and advanced RLVR environment
a06a840
verified

musharraf7 commited on

Sync with github: Training results and advanced RLVR environment
503bc84
verified

musharraf7 commited on

Sync with github: Training results and advanced RLVR environment
0d73a91
verified

musharraf7 commited on

Sync with github: Training results and advanced RLVR environment
f16c8fc
verified

musharraf7 commited on

chore: rename project to esctr-environment
b15694c

musharraf7 commited on

chore: add hf_token.txt to gitignore
361da2b

musharraf7 commited on

chore: update Space URL to esctr-environment
e15d46c

musharraf7 commited on

Delete course.md
381a57f
unverified

Shah Musharaf ul Islam commited on

feat: ESCTR pivot — Enterprise Supply Chain & Tax Reconciliation
a363048

musharraf7 commited on

feat: upgrade environment for hackathon finale
6f7e1b7

musharraf7 commited on

Delete project_juiding_criterion.txt
763c527
unverified

Shah Musharaf ul Islam commited on

Add 2 new frontier-challenging tasks + reward shaping system
a2ae67c

Musharraf commited on

Fix: clamp scores to strict (0,1) range - never 0.0 or 1.0
7de3176

Musharraf commited on