DataEngEnv / app

Commit History

  1. feat: semantic rewards on script state, raise step cap to 20, full-script rewrite support

    a0bbcba
  2. fix: align RewardEngine with actual stage tasks, fix step budget to 60, enable stage-specific reset

    aa04f19
  3. fix: wire RewardEngine into environment for dense rewards, fix step-limit mismatches

    a166934

  1. fix: track missing Codex files (mlops_bot, tasks, graders) and add app/actors/__init__.py to resolve ModuleNotFoundError

    eb71b92
  2. fix(round2): fix Stage 3 string matching in baseline script

    e91d5da
  3. feat(round2): add Stage 4 rewards and query_actor handling to RewardEngine

    8ee3fce

  1. feat(round2): build Stage 2 Training Monitor task, grader, and code reviewer bot

    42d30a9
  2. feat(round2): build Stage 1 Data Repair task and string-based deterministic grader

    051d18c

  1. fix: remove FastAPI root route to unblock Gradio UI on root path

    6a80f43
  2. fix: mount Gradio directly onto FastAPI to prevent HTTP 404 conflicts on evaluator routes

    edd5fd2
  3. fix: clamp all scores to strictly open interval (0.01, 0.99) to satisfy OpenEnv task validation

    0d0be12

  1. fix: rename entry point to main and add server/app.py for openenv multi-mode validation

    cb86053

  1. fix: satisfy openenv multi-mode package scaffolding requirements by resolving python dependencies array and bootstrapping UV lockfile constraints

    8828035

  1. fix: allow empty POST bodies for hackathon automated testing validation

    4ec4df3

  1. Add root endpoint for HF Spaces

    bd142b3
  2. DataEngEnv - Meta AI Hackathon submission

    4189f0e