Organize Open-SWE-Traces data prep project

2026-06-24 22:34:04 +08:00
commit f06e573b04
27 changed files with 3264 additions and 0 deletions
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -0,0 +1,29 @@
+[build-system]
+requires = ["setuptools>=68", "wheel"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "ti-coding-agent-data-prep"
+version = "0.1.0"
+description = "Open-SWE-Traces probing, filtering, and repurposing utilities for coding-agent SFT data preparation."
+readme = "README.md"
+requires-python = ">=3.10"
+dependencies = [
+  "datasets>=2.19.0",
+  "huggingface-hub>=0.23.0",
+  "pyarrow>=15.0.0",
+  "transformers>=4.52.0",
+]
+
+[project.optional-dependencies]
+dev = [
+  "ruff>=0.5.0",
+]
+
+[tool.setuptools.packages.find]
+where = ["src"]
+
+[tool.ruff]
+line-length = 120
+target-version = "py310"
+