{"success":true,"data":{"author":"Mirali Purohit, Hannah Kerner et al. (Mars-Bench, Arizona State University)","bucketAccessType":"public","bucketName":"Mirali33/mb-surface_multi_label_cls","bucketPrefix":"","category":"Astronomy & Space","cloudProvider":"huggingface","createdAt":"2026-11-18T00:06:03.760Z","description":"Multi-label classification of 1024 x 1024 Pancam images from the Mars Exploration Rovers Spirit and Opportunity, revised from Cole et al. (2020). Each image can carry several of 25 labels describing geology and rover context: soil, bright soil, soil trench, rock outcrops, float rocks, clasts, spherules, dunes and ripples, rocks with round or linear features, close-up rock, Rock Abrasion Tool (RAT) holes and brushed targets, distant vista, sky, astronomy, rover tracks, rover deck and parts, arm hardware and the Pancam calibration target. Splits: 1,762 training, 443 validation and 739 test images, with few-shot subsets. Parquet shards under data/ with image and label list, plus annotation.csv and mapping.json; load with datasets.load_dataset(\"Mirali33/mb-surface_multi_label_cls\"). About 1.2 GB. Part of Mars-Bench (Purohit et al., NeurIPS 2025 Datasets and Benchmarks track, arXiv:2510.24010), 20 standardized Mars datasets released by the authors under CC BY 4.0; cite Mars-Bench and Cole et al. (2020).","displayTitle":"Mars-Bench MER Pancam Multi-Label Surface Classification (Spirit and Opportunity, 25 labels)","doi":"10.48550/arXiv.2510.24010","huggingFaceUrl":"https://huggingface.co/datasets/Mirali33/mb-surface_multi_label_cls","id":"e1cf9c53-d965-5b16-b8d3-28335d95846a","license":"CC-BY-4.0","licenseEvidenceUrl":"https://arxiv.org/html/2510.24010v1","objectCountEstimate":20,"probeStatus":"ok","referenceUrl":"https://huggingface.co/datasets/Mirali33/mb-surface_multi_label_cls","region":"us","schemaCard":{"accessHints":[{"language":"python-datasets","snippet":"from datasets import load_dataset\nds = load_dataset(\"Mirali33/mb-surface_multi_label_cls\")"},{"language":"python-hub","snippet":"from huggingface_hub import snapshot_download\nsnapshot_download(repo_id=\"Mirali33/mb-surface_multi_label_cls\", repo_type=\"dataset\")"}],"caveats":["HuggingFace metadata: author=Mirali33, license:cc-by-4.0, size_categories:1K<n<10K, format:parquet, downloads=71"],"format":"mixed","generatedAt":"2026-11-18T00:09:08.166Z","layout":{"example":"annotation.csv","pattern":"{repoId}/{filename}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"file_id,split,feature_name,label\n1p128287181mrd0000p2303l2m1.jpg,train,\"['soi', 'sph', 'sky', 'nbs']\",\"[3, 15, 21, 18]\"\n1p128287268mrd0000p2303l2m1.jpg,train,\"['rod', 'dur', 'soi', 'sph', 'nbs', 'rpa']\",\"[8, 2, 3, 15, 18, 22]\"\n1p128288415mrd0000p2303l2m1.jpg,train,\"['rmi', 'dur', 'soi', 'sph', 'sky', 'nbs']\",\"[19, 2, 3, 15, 21, 18]\"\n1p128289213mrd0000p2303l2m1.jpg,train,\"['soi', 'sph', 'nbs']\",\"[3, 15, 18]\"\n1p129341281mrd0300p2376l2m1.jpg,train,\"['roc', 'cla', 'rlf', 'soi', 'sph', 'nbs']\",\"[4, 1, 23, 3, 15,","encoding":"utf8","key":"annotation.csv","truncatedAt":512}],"sampleObjects":["annotation.csv","data/few_shot_train_10_shot-00000-of-00001.parquet","data/few_shot_train_15_shot-00000-of-00001.parquet","data/few_shot_train_1_shot-00000-of-00001.parquet","data/few_shot_train_20_shot-00000-of-00001.parquet"],"sidecars":[{"bytes":"---\nannotations_creators:\n- expert-generated\nlanguage_creators:\n- found\nlanguage:\n- en\nlicense:\n- cc-by-4.0\nmultilinguality:\n- monolingual\nsize_categories:\n- 1K<n<10K\nsource_datasets:\n- original\ntask_categories:\n- image-classification\ntask_ids:\n- multi-label-image-classification\npretty_name: MER - Mars Exploration Rover Dataset\n---\n\n# MER - Mars Exploration Rover Dataset\n\nA multi-label classification dataset containing Mars images from the Mars Exploration Rover (MER) mission for planetary science research.\n\n## Dataset Metadata\n\n* **License:** CC-BY-4.0 (Creative Commons Attribution 4.0 International)\n* **Version:** 1.0\n* **Date Published:** 2025-10-23\n* **Cite As:** TBD\n\n## Classes\n\nThis dataset uses multi-label classification, meaning each image can have multiple class labels.\n\nThe dataset contains the following classes:\n\n- **rah** (0): Rock Abrasion Tool (RAT) Hole\n- **cla** (1): Clasts\n- **dur** (2): Dunes/Ripples\n- **soi** (3): Soil\n- **roc** (4): Rock Outcrops\n- **clr** (5): Close-up Rock\n- **rab** (6): Rock Abrasion Tool (RAT) Brushed Target\n- **div** (7): Distant Vista\n- **rod** (8): Rover Deck\n- **bso** (9): Bright Soil\n- **flr** (10): Float Rocks\n- **art** (11): Artifacts\n- **pct** (12): Pancam Calibration Target\n- **arh** (13): Arm Hardware\n- **rrf** (14): Rock (Round Features)\n- **sph** (15): Spherules\n- **ohw** (16): Other Hardware\n- **ast** (17): Astronomy\n- **nbs** (18): Nearby Surface\n- **rmi** (19): Rocks (Misc)\n- **rtr** (20): Rover Tracks\n- **sky** (21): Sky\n- **rpa** (22): Rover Parts\n- **rlf** (23): Rock (Linear Features)\n- **sot** (24): Soil Trench\n## Statistics\n\n- **train**: 1762 images\n- **val**: 443 images\n- **test**: 739 images\n- **few_shot_train_10_shot**: 128 images\n- **few_shot_train_15_shot**: 175 images\n- **few_shot_train_1_shot**: 16 images\n- **few_shot_train_20_shot**: 220 images\n- **few_shot_train_2_shot**: 30 images\n- **few_shot_train_5_shot**: 67 images\n- **partition_0.01x_partition**: 19 images\n- **partition_0.02x_partition**: 33 images\n- **partition_0.05x_partition**: 81 images\n- **partition_0.10x_partition**: 184 images\n- **partition_0.20x_partition**: 361 images\n- **partition_0.25x_partition**: 447 images\n- **partition_0.50x_partition**: 878 images\n\n## Few-shot Splits\n\nThis dataset includes the following few-shot training splits:\n\n- **few_shot_train_10_shot**: 128 images\n- **few_shot_train_15_shot**: 175 images\n- **few_shot_train_1_shot**: 16 images\n- **few_shot_train_20_shot**: 220 images\n- **few_shot_train_2_shot**: 30 images\n- **few_shot_train_5_shot**: 67 images\n\nFew-shot configurations:\n\n- **10_shot.csv**\n- **15_shot.csv**\n- **1_shot.csv**\n- **20_shot.csv**\n- **2_shot.csv**\n- **5_shot.csv**\n## Partition Splits\n\nThis dataset includes the following partition splits:\n\n- **partition_0.01x_partition**: 19 images\n- **partition_0.02x_partition**: 33 images\n- **partition_0.05x_partition**: 81 images\n- **partition_0.10x_partition**: 184 images\n- **partition_0.20x_partition**: 361 images\n- **partition_0.25x_partition**: 447 images\n- **partition_0.50x_partition**: 878 images\n\nPartition configurations:\n\n- **0.01x_partition.csv**\n- **0.02x_partition.csv**\n- **0.05x_partition.csv**\n- **0.10x_partition.csv**\n- **0.20x_partition.csv**\n- **0.25x_partition.csv**\n- **0.50x_partition.csv**\n## Format\n\nEach example in the dataset has the following format:\n\n```\n{\n  'image': Image(...),  # PIL image\n  'label': List[int],  # Multi-hot encoded binary vector (1 if class is present, 0 otherwise)\n  'feature_name': List[str],  # List of feature names (class short codes)\n}\n```\n\n## Usage\n\n```python\nfrom datasets import load_dataset\n\ndataset = load_dataset(\"Mirali33/mb-surface_multi_label_cls\")\n\n# Access an example\nexample = dataset['train'][0]\nimage = example['image']  # PIL image\nlabel = example['label']  # Multi-hot encoded binary vector\n\n# Example of how to find which classes are present in an image\npresent_classes = [i for i, is_present in enumerate(label) if is_present == 1]\nprint(f\"Classes present in this image: {present_classes}\")\n```\n\n## Multi-label Classification\n\nIn multi-label classification, each","encoding":"utf8","key":"README.md","name":"README.md","size":4501,"truncated":true}],"sourceObjectKey":"annotation.csv","starterCell":"import pandas as pd\ndf = pd.read_csv(\"hf://datasets/Mirali33/mb-surface_multi_label_cls/annotation.csv\")\ndf.head()","starterSamplePath":"annotation.csv","totalObjects":20},"summary":"Multi-label classification of 1024 x 1024 Pancam images from the Mars Exploration Rovers Spirit and Opportunity, revised from Cole et al. (2020).","title":"Mars-Bench MER Pancam Multi-Label Surface Classification (Spirit and Opportunity, 25 labels)","updatedAt":"2026-11-18T03:55:59.297Z","visibility":"public","websiteUrl":"https://mars-bench.github.io/"}}