codefix-env/openenv.yaml at main · dhakarshailendra829/codefix-env · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
name: codefix-env
version: "0.2.0"
description: >
  A high-performance PyTorch-based RL environment for automated Python code
  debugging and LLM agent evaluation. Built on Meta OpenEnv / Gymnasium API.

author: Shailendra Dhakar
license: MIT
repository: https://github.com/dhakarshailendra829/codefix-env

# OpenEnv compatibility
openenv_version: ">=0.1.0"
api_style: gymnasium

# Server config
server:
  module: server.app
  app: app
  host: "0.0.0.0"
  port: 8000
  workers: 2

# Environment metadata
environment:
  action_type: CodeFixAction
  observation_type: CodeFixObservation
  state_type: CodeFixState
  max_steps: 20
  episode_based: true

# Available tasks
tasks:
  easy:
    count: 8
    categories: [syntax_error, logic_bug, off_by_one, type_error, return_bug, wrong_operator, index_error]
  medium:
    count: 8
    categories: [algorithm_bug, recursion_bug, scope_bug, mutable_default, logic_bug]
  hard:
    count: 5
    categories: [multi_function, lru_cache, graph_algorithm, decorator, data_structures]

# Dependencies
requires:
  python: ">=3.10"
  packages:
    - fastapi>=0.110.0
    - pydantic>=2.6.0
    - torch>=2.2.0
    - httpx>=0.27.0
    - structlog>=24.1.0
    - rich>=13.7.0

# Deployment
docker:
  image: ghcr.io/dhakarshailendra829/codefix-env
  dockerfile: server/Dockerfile
  port: 8000

huggingface:
  space: dhakarshailendra829/codefix-env
  sdk: docker

tags:
  - reinforcement-learning
  - code-debugging
  - pytorch
  - llm
  - gymnasium
  - openenv
  - genai
  - software-engineering