format_version int64 | is_processed bool | num_samples int64 | voxel_shape list | voxel_dtype string | paths dict | class_labels_format string | num_blocks int64 | num_classes int64 | metadata_applied_during_processing bool | source dict | sample_of string | sample_source_indices_path string | sample_selection dict |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2 | true | 2,688 | [
32,
32,
32
] | int32 | {
"voxels": "voxels.npy",
"biome_labels": "biome_labels.npy"
} | indices | 177 | 21 | false | {
"path": "data/natural_human_balanced_dataset_cleaned",
"had_metadata": false,
"metadata_applied_at_source": true
} | human_augmented_dataset/natural_human_balanced_dataset_cleaned_cc_val_dir | source_indices.npy | {
"strategy": "deterministic stratified sample by integer label",
"per_label_target": 128,
"selected_source_indices_count": 2688,
"labels": {
"0": {
"available": 4818,
"sampled": 128
},
"1": {
"available": 4765,
"sampled": 128
},
"2": {
"available": 4811,
... |
Dream-Cubed Human Representative Sample
This directory is a deterministic, class-stratified sample of the Dream-Cubed Human dataset. It is provided for reviewer inspection of data quality; the full dataset remains available at https://huggingface.co/datasets/dream-cubed/DreamCubedHumanSample.
The sample dataset for procedurally generated data is available at https://huggingface.co/datasets/dream-cubed/DreamCubedNatural.
Depending on which commands have been run, the sample may include raw human-authored map chunks and/or compact processed train/validation examples. Processed examples may also contain natural biome chunks from the companion Dream-Cubed Natural release. Permission to release voxelized chunks from the human-authored maps was granted explicitly by the anonymized map creator for noncommercial research, but not for commercial products or reconstruction of the original maps.
Creation Procedure
- Sampling seed:
67 - Last source mode added:
raw - Raw target: up to
32chunks per raw source label - Processed target: up to
128chunks per label per split - Strategy: deterministic stratified sampling without replacement
- Output: compact
.npzraw samples and restored, unsharded processed.npysplit samples
Exact source paths, selected source indices, sample counts, and output files are recorded in
sample_manifest.json. Processed split directories also include source_indices.npy and fresh
manifest.json files. Dataset metadata files copied from nearby public dataset files are stored
under metadata/ when available.
Inspection Notes
This sample is intentionally small and should not be used for model training or quantitative evaluation. It is meant only to help reviewers inspect file formats, labels, chunk tensors, and overall data quality without downloading the full multi-GB dataset.
- Downloads last month
- 8