Datasets:
Tasks:
Text Generation
Formats:
csv
Sub-tasks:
language-modeling
Languages:
English
Size:
10K - 100K
License:
File size: 1,396 Bytes
bd5f048 | 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 | SWEGraph v3 — FREE SAMPLE
Copyright (c) 2026 williamTLmiller (energea@energeatech.com)
WHAT THE CC BY-NC 4.0 LICENSE IN THIS REPOSITORY COVERS
Only williamTLmiller's normalization and packaging work: the proof-chain
tables, identity and contamination layer, license detection, environment
links, task validity records and execution evidence, split assignment,
provenance manifests and documentation. Non-commercial use, with
attribution.
WHAT IT DOES NOT COVER
The underlying releases this is built from — SWE-bench, SWE-bench Verified,
Lite, Multimodal and Multilingual (MIT), SWE-smith (MIT), SWE-bench Pro
(MIT tooling) and BigCodeBench (Apache-2.0) — and the open-source
repository content they carry, which stays under each repository's own
license. Copies of the source licenses are included as
LICENSE-*-UNDERLYING-*.txt. Nothing here restricts what you may do with
those sources obtained from their publishers.
NOT AFFILIATED
Not produced, reviewed or endorsed by the authors of any source dataset.
ATTRIBUTION
"SWEGraph v3 sample, MatrixNorm normalization by williamTLmiller, derived
from SWE-bench and SWE-smith."
(This free sample holds SWE-bench train and SWE-smith tasks only; the full
SWEGraph v3 package also draws on SWE-bench Pro and BigCodeBench.)
The full dataset is available under a separate commercial license — see README.md.
|