Croissant: v1.1.0, refreshed checksums, correct geophys channel count

croissant.json

@@ -45,14 +45,14 @@
   "description": "First benchmark for data-driven annual calving front forecasting, comprising 17,358 observations across 123 marine-terminating Svalbard glaciers from 2013 to 2023.",
   "url": "https://huggingface.co/datasets/enscg/calvdb",
   "license": "https://creativecommons.org/licenses/by/4.0/",
-  "version": "1.0.0",
-  "datePublished": "2026-05-01",
+  "version": "1.1.0",
+  "datePublished": "2026-07-26",
   "conformsTo": "http://mlcommons.org/croissant/1.0",
-  "rai:dataLimitations": "The dataset is restricted to 123 marine-terminating glaciers in the Svalbard archipelago (RGI 6.0, region 07) and covers only 2013–2023. The prediction task is fixed to an annual (365-day) horizon; sub-annual forecasting is out of scope. The dataset focuses on calving-front position as the target variable, so other aspects of glacier dynamics (e.g., surface velocity, mass balance) are not represented.",
+  "rai:dataLimitations": "The dataset is restricted to 123 marine-terminating glaciers in the Svalbard archipelago (RGI 6.0, region 07) and covers only 2013\u20132023. The prediction task is fixed to an annual (365-day) horizon; sub-annual forecasting is out of scope. The dataset focuses on calving-front position as the target variable, so other aspects of glacier dynamics (e.g., surface velocity, mass balance) are not represented.",
   "rai:dataBiases": "Geographic bias: all 123 glaciers are located in Svalbard, which has a distinct climate regime, fjord geometry, and ocean-forcing characteristics compared with other Arctic and Antarctic calving systems; models trained solely on this dataset may not transfer to other regions without fine-tuning.",
   "rai:personalSensitiveInformation": "This dataset contains no personal or sensitive information. All data are derived entirely from satellite remote-sensing imagery, geophysical measurements, and climate reanalysis products. No information relating to human subjects is present.",
   "rai:dataUseCases": "The dataset is designed to measure a model's ability to predict the spatial location of a marine-terminating glacier's calving front 365 days in the future, given a sequence of five past calving-front locations together with co-registered multispectral imagery, geophysical covariates, and climate forcing.",
-  "rai:dataSocialImpact": "The dataset enables data-driven monitoring of glacier calving dynamics, advancing scientific understanding of ice–ocean interactions and sea-level rise projections relevant to climate-change adaptation and coastal planning. Releasing the benchmark publicly lowers the barrier for the machine-learning and remote-sensing communities to engage with cryosphere science.",
+  "rai:dataSocialImpact": "The dataset enables data-driven monitoring of glacier calving dynamics, advancing scientific understanding of ice\u2013ocean interactions and sea-level rise projections relevant to climate-change adaptation and coastal planning. Releasing the benchmark publicly lowers the barrier for the machine-learning and remote-sensing communities to engage with cryosphere science.",
   "rai:hasSyntheticData": false,
   "prov:wasDerivedFrom": "Full details of data sources are provided in Section 3 (Dataset) of the accompanying paper.",
   "prov:wasGeneratedBy": "Full details of data collection, preprocessing, annotation, and benchmark construction are provided in Section 4 (CalvingDB Benchmark) of the accompanying paper.",
@@ -64,7 +64,7 @@
       "description": "Training split sample index (88 glaciers).",
       "contentUrl": "splits/train.json",
       "encodingFormat": "application/json",
-      "sha256": "6e3fa98028a5ec6ec9fdf56350310b297338593db9f8437bd53ab92d890cc658"
+      "sha256": "458f21f24c30299d94fc5643b1ce6213eb1bb8745a148f92fb78890f747cc96b"
     },
     {
       "@type": "cr:FileObject",
@@ -73,7 +73,7 @@
       "description": "Validation split sample index (12 glaciers).",
       "contentUrl": "splits/val.json",
       "encodingFormat": "application/json",
-      "sha256": "7de1bb38c0b0c7f50f623129923da92add572f11287ca68a442a34e62e3e4325"
+      "sha256": "30783734b1222c965fda36409d9a369b5a79319d09e90824b6ab04f315480df8"
     },
     {
       "@type": "cr:FileObject",
@@ -82,7 +82,7 @@
       "description": "Test split sample index (23 glaciers).",
       "contentUrl": "splits/test.json",
       "encodingFormat": "application/json",
-      "sha256": "e6a06da202207c9f14b6010e720be2723f6750a4e89227385e16477151977788"
+      "sha256": "c6bf8e469e72821e832371dacd5517c498b08980d50564b96b6691e662f35a83"
     },
     {
       "@type": "cr:FileObject",
@@ -97,9 +97,9 @@
       "@type": "cr:FileSet",
       "@id": "zarr-stores",
       "name": "zarr_stores",
-      "description": "Per-glacier zarr stores compressed as zip files. Each zip contains arrays: sdt (T,H,W), imagery (T,5,H,W), geophys (T,5,H,W), climate (T,2), trace (T,H,W). Spatial resolution: 30m, CRS: EPSG:3995.",
+      "description": "Per-glacier Zarr v3 stores, each packaged as a zip archive readable directly with zarr.storage.ZipStore (no unzipping required). Arrays: sdt (T,H,W), trace (T,H,W), imagery (T,5,H,W), geophys (T,6,H,W), climate (T,2), dates (T,), sensors (T,), and the validity flags valid_sdt/valid_img/valid_clm/valid_trace (T,). All arrays carry dimension_names. Spatial resolution 30 m, CRS EPSG:3995. Requires zarr-python >= 3.0.",
       "encodingFormat": "application/zip",
       "includes": "zarr_zipped/*.zip"
     }
   ]
-}
+}