Datasets:
Sync Croissant URL to anonymous.4open.science
Browse files- CROISSANT.json +1 -1
CROISSANT.json
CHANGED
|
@@ -40,7 +40,7 @@
|
|
| 40 |
"description": "Synthetic time-series datasets with mathematically exact ground-truth labels for the TSFMI baseline-controlled evaluation protocol. Six standard properties (trend, seasonality, frequency, stationarity, anomaly, change point) plus five hard variants. Each dataset is generated on-demand by deterministic seeded generators in src/datasets/synthetic.py; this manifest documents the canonical seed=42 instantiation used for all confirmatory results in the paper. The TSFMI paper itself is the protocol/benchmark contribution; the datasets here exist only to instantiate the protocol and have no claim of independent novelty.",
|
| 41 |
"conformsTo": "http://mlcommons.org/croissant/1.0",
|
| 42 |
"license": "https://opensource.org/licenses/MIT",
|
| 43 |
-
"url": "https://anonymous.4open.science/r/
|
| 44 |
"version": "1.0.0",
|
| 45 |
"datePublished": "2026-05-06",
|
| 46 |
"isLiveDataset": false,
|
|
|
|
| 40 |
"description": "Synthetic time-series datasets with mathematically exact ground-truth labels for the TSFMI baseline-controlled evaluation protocol. Six standard properties (trend, seasonality, frequency, stationarity, anomaly, change point) plus five hard variants. Each dataset is generated on-demand by deterministic seeded generators in src/datasets/synthetic.py; this manifest documents the canonical seed=42 instantiation used for all confirmatory results in the paper. The TSFMI paper itself is the protocol/benchmark contribution; the datasets here exist only to instantiate the protocol and have no claim of independent novelty.",
|
| 41 |
"conformsTo": "http://mlcommons.org/croissant/1.0",
|
| 42 |
"license": "https://opensource.org/licenses/MIT",
|
| 43 |
+
"url": "https://anonymous.4open.science/r/TSFMI/",
|
| 44 |
"version": "1.0.0",
|
| 45 |
"datePublished": "2026-05-06",
|
| 46 |
"isLiveDataset": false,
|