AstroFetch

PyTorch-friendly, ML-ready access to planetary science data, starting with the Moon. Request a bounding box, receive a coregistered multichannel tensor.

Phase 1 (STAC sampler)

The data path is real: a request fetches the Cloud Optimized GeoTIFFs covering the window from the USGS ARD catalog, reprojects them onto a common grid, and returns physical values. See the roadmap.

Install

uv add astrofetch      # or: pip install astrofetch

Quickstart

One dataset class per instrument; combine instruments with & to stack their channels over the overlapping region:

import astrofetch as af

bbox = (-26.3, -50.6, -25.5, -49.7)  # west, south, east, north (degrees)

moondata = af.KaguyaTC(products=["dtm"], bbox=bbox, resolution=100) & af.KaguyaTCImagery(
    bbox=bbox, resolution=100
)

for sample in moondata:
    sample["image"]   # torch.Tensor (C, H, W), one channel per layer
    sample["mask"]    # torch.BoolTensor (C, H, W), validity (nodata gaps)
    sample["layers"]  # ["kaguya_tc_dtm", "kaguya_tc_image"], plus bbox/crs/resolution

Samples are plain dicts, so a DataLoader collates them with no custom code:

from torch.utils.data import DataLoader

loader = DataLoader(moondata, batch_size=16)
for batch in loader:
    batch["image"]  # torch.Tensor (16, C, H, W)

Beyond the STAC catalog

Some instruments (LROC, LOLA, Mini-RF, Diviner, ShadowCam, Clementine, ...) aren't in the USGS ARD STAC catalog at all; those datasets search the NASA PDS Orbital Data Explorer instead, or read a single fixed mosaic URL, behind the exact same interface. The full roster is in Instrument datasets; two examples:

import astrofetch as af

# LROC NAC stereo DTM sites are a few hundred named sites, not global
# coverage, so sampled windows are drawn from inside a real site by default.
nac = af.LROCNACDTM(products=["dtm", "ortho"], bbox=(3.0, 25.0, 4.5, 26.5))
sample = nac[0]

# A global 100 m WAC mosaic and a global LOLA DEM, channel-stacked with `&`.
terrain = af.LROCWACMosaic(resolution=100) & af.LOLA(resolution=100)

# ShadowCam mosaics of permanently shadowed polar craters, another
# site-based instrument -- footprint sampling applies the same way.
shadowcam = af.ShadowCam(products=["mosaic"])

Discovering what data exists

The MOON catalog enumerates probes, instruments, and products, and points at the dataset classes:

from astrofetch.moon import MOON

for probe in MOON.probes.values():
    for instrument in probe.instruments.values():
        print(probe.name, "/", instrument.name, "->", sorted(instrument.products))

For the full design rationale and contributor guidance, see AGENTS.md.