Files
OSGKeyboard/ModelTraining/ClipboardSemantics/online-real-holdout-sources.json
T
2026-08-27 18:01:46 +08:00

43 lines
1.5 KiB
JSON

{
"containsProjectUserClipboardData": false,
"corpusSHA256": "33d5fdfb3dc2d3f758cc9f0b1878c90a4a01203a7d8ce77c0082969bfdb5c4c3",
"generatedFromOnlineSources": true,
"labelCorrection": {
"reason": "Replaced generic MASSIVE device-control commands with attributed GitHub work requests.",
"supersededCorpusSHA256": "2962b6d28fe0ef551ae46f62accbde93c9da7f0fad93451c06c2e226350aad33"
},
"labelDefinition": {
"complaint": "A user reports a negative experience, failure, or grievance.",
"task": "An actionable work request suitable for todo extraction; device-control commands are excluded."
},
"recordCount": 200,
"seed": 20260831,
"sources": [
{
"dataset": "CFPB via hpe-ai/customer-complaints",
"license": "CC0-1.0 source / Apache-2.0 mirror",
"records": 40,
"url": "https://huggingface.co/datasets/hpe-ai/customer-complaints"
},
{
"dataset": "Meituan-Dianping/ASAP",
"license": "Apache-2.0",
"records": 40,
"revision": "975122a60065240124df62cb4d5dbfd19ed9ef2c",
"url": "https://github.com/Meituan-Dianping/asap"
},
{
"dataset": "AmazonScience/MASSIVE",
"license": "CC-BY-4.0",
"records": 80,
"url": "https://huggingface.co/datasets/AmazonScience/massive"
},
{
"dataset": "Sulak2020 GitHub issues multi-repo dataset",
"license": "CC-BY-4.0 with per-record attribution",
"records": 40,
"url": "https://huggingface.co/datasets/Sulak2020/github-issues-multirepo-datasets-Sulakshana"
}
]
}