Skip to content

CLAUDE.md

Latest commit

 

History

History
40 lines (31 loc) · 1.59 KB

CLAUDE.md

File metadata and controls

40 lines (31 loc) · 1.59 KB

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

Purpose

This repository sets up a Jupyter-based Docker environment for learning and experimenting with geospatial/scientific Python libraries:

  • xarray — labeled N-dimensional arrays
  • zarr — chunked, compressed array storage
  • pystac-client / planetary-computer — STAC catalog access (e.g., Microsoft Planetary Computer)
  • rioxarray — rasterio extension for xarray
  • odc-stac — Open Data Cube STAC utilities
  • dask[complete] — parallel/distributed computing
  • ipywidgets — interactive Jupyter widgets

Building and Running

# Build the image
docker build -t xarray-learn .

# Run (bash — Git Bash / WSL / Mac / Linux)
docker run -p 8888:8888 \
  -v "$PWD":/home/jovyan/work \
  -v "$USERPROFILE/.aws":/home/jovyan/.aws:ro \
  xarray-learn

# Run (cmd.exe)
# docker run -p 8888:8888 -v "%cd%":/home/jovyan/work -v "%USERPROFILE%\.aws":/home/jovyan/.aws:ro xarray-learn

Project rules

  • Use xarray + Zarr for all storage/analysis (chunked, dask-backed).
  • Prefer pystac-client + odc-stac or rioxarray for STAC → Xarray.
  • Always explain library concepts when making changes.
  • Run notebook cells via jupyter nbconvert --execute or papermill after edits and fix failures.
  • Verify Zarr roundtrips with xarray.open_zarr().
  • Specification-driven: never code without referencing SPEC.md.

The base image comment references jupyter/scipy-notebook or similar — update the FROM line at the top of the Dockerfile to specify the actual base image before building.