calebrob6

#!/usr/bin/env python3
"""Download a Sentinel-2 time series from the Microsoft Planetary Computer.

Companion script for the blog post "Super-Resolving Sentinel-2 with Gaussian
Splats" (https://geospatialml.com/posts/sentinel2-superresolution/).

Pulls N cloud-free Sentinel-2 L2A scenes over a configurable AOI, crops to a
common pixel window, and writes one 4-band GeoTIFF per scene (B02 Blue,
B03 Green, B04 Red, B08 NIR; uint16 reflectance in 0..10000).

calebrob6

#!/usr/bin/env python3
"""Benchmark iteration over the IOBench dataset.

For each requested split (``raw`` and/or ``preprocessed``), this script builds
an :class:`~torchgeo.datasets.IOBench` dataset, wraps it in a
:class:`~torchgeo.samplers.GridGeoSampler` (non-overlapping patches), feeds it
through a :class:`torch.utils.data.DataLoader`, and reports timing information
for one full epoch.

Example usage::

calebrob6

calebrob6

#!/usr/bin/env python3
"""Compute mean and stdev of Alpha Earth Foundation embeddings per census block.

Required input data
-------------------
1. Census block shapefiles:
   - Go to https://www.census.gov/cgi-bin/geo/shapefiles/index.php
   - Select year "2025" (or desired year) and layer type "Blocks (2020)"
   - Choose a state (e.g. "Washington") and download the ZIP
   - Unzip to get tl_2025_53_tabblock20.shp (and companion files)

calebrob6

calebrob6

calebrob6

calebrob6

import argparse
import math
import os
import time
from typing import List, Optional, Sequence, Tuple

import numpy as np
import rasterio
import rasterio.windows
import torch

calebrob6

calebrob6

def batch_histogram(data_tensor, num_classes=-1):
    """
    From https://discuss.pytorch.org/t/batched-torch-histc/179741
    
    Computes histograms of integral values, even if in batches (as opposed to torch.histc and torch.histogram).
    Arguments:
        data_tensor: a D1 x ... x D_n torch.LongTensor
        num_classes (optional): the number of classes present in data.
                                If not provided, tensor.max() + 1 is used (an error is thrown if tensor is empty).
    Returns: