CoolFace
Datasetpublic

SuperComputer/mars_hirise_dtm_raw-42e7c107a4f7c2d6

Mars HiRISE DTM — LitData Streaming Dataset Pre-processed Mars HiRISE orthoimage + DTM patches in LitData optimized streaming format. Associated with MarsFM: Shading-Regularized Flow Matching for Martian Relief Estimation. Quick Start from litdata import StreamingDataset # Stream directly from HuggingFace — no full download needed train_ds = StreamingDataset(input_dir="hf://datasets/SuperComputer/mars_hirise_dtm_raw-42e7c107a4f7c2d6/train") val_ds =… See the full description on the dataset page: https://huggingface.co/datasets/SuperComputer/mars_hirise_dtm_raw-42e7c107a4f7c2d6.

sourceHugging Faceapache-2.0updated 2d agoView on Hugging Face
0likes540downloads
Dataset Card

Mars HiRISE DTM — LitData Streaming Dataset

Pre-processed Mars HiRISE orthoimage + DTM patches in LitData optimized streaming format. Associated with MarsFM: Shading-Regularized Flow Matching for Martian Relief Estimation.

Quick Start

python
from litdata import StreamingDataset

# Stream directly from HuggingFace — no full download needed
train_ds = StreamingDataset(input_dir="hf://datasets/SuperComputer/mars_hirise_dtm_raw-42e7c107a4f7c2d6/train")
val_ds   = StreamingDataset(input_dir="hf://datasets/SuperComputer/mars_hirise_dtm_raw-42e7c107a4f7c2d6/val")
test_ds  = StreamingDataset(input_dir="hf://datasets/SuperComputer/mars_hirise_dtm_raw-42e7c107a4f7c2d6/test")

sample = train_ds[0]

# Core Tensors
print(sample["elevation"].shape)    # (1, H, W) float32
print(sample["bounds"].shape)       # (4,)      float32
print(sample["crs"])                # str

# Orthoimages (presence depends on config)
if "left_red" in sample:
    print(sample["left_red"].shape) # (1, H, W) float32
if "left_irb" in sample:
    print(sample["left_irb"].shape) # (3, H, W) float32

# Flattened Metadata
print(sample["dtm_product_id"])     # str
print(sample["left_obs_id"])        # str
print(sample["incidence_angle"])    # float32

Preprocessing Configuration

Config Hash: 42e7c107a4f7c2d6

This dataset was generated with the following pipeline parameters:

yaml
hirise:
  root: /scratch/mars_hirise_dtm
  include_ortho: true
  ortho_type: RED
  ortho_scale: null
  download: false
  reuse_cache: true
  target: null
  bbox:
  - 60
  - 0
  - 90
  - 30
sampler:
  size: 0.018
  length: null

(Note: Depending on the dataset configuration, some orthoimage keys may be omitted).