-
Notifications
You must be signed in to change notification settings - Fork 65
Expand file tree
/
Copy path.gitignore
More file actions
136 lines (122 loc) · 2.44 KB
/
Copy path.gitignore
File metadata and controls
136 lines (122 loc) · 2.44 KB
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
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
# Byte-compiled / optimized / DLL files
__pycache__/
*.py[cod]
*$py.class
*debug*/
# C extensions
*.so
# Distribution / packaging
/.Python
/build/
/dist/
/downloads/
/eggs/
/.eggs/
/lib/
/lib64/
/parts/
/sdist/
/var/
/wheels/
/share/python-wheels/
*.egg-info/
.installed.cfg
*.egg
MANIFEST
# Unit test / coverage reports
htmlcov/
.tox/
.nox/
.coverage
.coverage.*
.cache
nosetests.xml
coverage.xml
*.cover
*.py,cover
.hypothesis/
.pytest_cache/
cover/
# Type check / lint caches
.mypy_cache/
.dmypy.json
dmypy.json
.pyre/
.pytype/
.ruff_cache/
# Environments
.env
.venv
env/
venv/
ENV/
env.bak/
venv.bak/
!unirl/rollout/env/
!unirl/rollout/env/**/
!unirl/rollout/env/**/*.py
!unirl/rollout/env/**/*.md
# Jupyter / notebooks
.ipynb_checkpoints/
# Logs / experiment outputs
wandb/
outputs/
logs/
local/
!unirl/reward/local/
**/rollout_data/
**/buffer_stats/
*.out
*.pkl
*.log
# Datasets: converter code, guides, and committed prompt sets are tracked;
# everything else under datasets/ is downloaded or generated and stays local.
# Keyed on file type rather than on dataset name, so adding a new
# datasets/<name>/ needs no entry here, and a converter whose --out-dir points
# inside datasets/ (several guides cook that way) cannot leak a cooked dataset
# into the tree.
datasets/**
!datasets/**/
!datasets/**/*.py
!datasets/**/*.md
# Committed prompt sets: small versioned inputs, not generated data. Committing
# data is the one case that should need an explicit decision here.
!datasets/geneval/*.jsonl
!datasets/geneval2/synthetic/*.jsonl
!datasets/image_edit/*.jsonl
!datasets/ocr/*.txt
!datasets/pickscore/*.txt
# Downloaded snapshots and cooked outputs, whatever they happen to contain --
# last, so the code re-includes above do not reach inside them.
datasets/**/raw/
datasets/**/processed/
# Default --out-dir of the datasets/ converters
/data/
# Large model artifacts (defense-in-depth)
models/**/*.bin
models/**/*.pt
models/**/*.pth
models/**/*.ckpt
models/**/*.safetensors
# IDE / editor metadata
.vscode/
.idea/
settings.json
*.iml
*.code-workspace
.DS_Store
.claude/*
!.claude/skills/
.cursor/*
!.cursor/skills/
!.cursor/rules/
REFACTOR_LOG.md
uv.lock
# Benchmarks: fetched / gated data and results (see benchmarks/README.md)
benchmarks/image/dpg_bench/data/
benchmarks/text/aime/data/aime2025.jsonl
benchmarks/text/gpqa/data/
benchmarks_results/
# Private experimental packages (dirs or stray files) stay local-only
# (the tier is also not packaged).
experimental/private_*