NeuralGCM is a research-grade hybrid machine learning and physics-based atmospheric model for weather and climate simulation, implemented primarily in JAX and Haiku. The repository provides a well-structured library (`neuralgcm`) with a clean inference API (`PressureLevelModel`) that supports loading pretrained checkpoints, encoding ERA5-like data into the model's sigma-coordinate state, advancing the model in time, and decoding back to pressure-level output with xarray integration. The code is highly modular and configurable via the gin-config system, allowing flexible composition of encoders, decoders, advanced neural parameterization modules, stochastic random fields, and data transforms. The repository also includes reference training code (designated as pseudocode), paper configuration files, Jupyter notebooks demonstrating data preparation and model deep dives, and pre-trained deterministic and stochastic checkpoints at multiple resolutions. While the inference path is fully reproducible given the provided checkpoints and public ERA5 data, the training pipeline relies on internal Google infrastructure and proprietary code, limiting full reproducibility. The library demonstrates strong software engineering practices for a research codebase, with extensive unit tests, modular design, and clear versioning, though it exhibits some technical drift due to an ongoing migration of components to a separate repository (`terrax`) and the use of hardcoded normalization statistics tied to specific training datasets.
wrf-python is a specialized Python library for computing over 30 diagnostic meteorological variables from WRF-ARW (Weather Research and Forecasting) model output. It provides a high-level `getvar()` interface that dispatches to Fortran and C computational kernels wrapped via f2py, with OpenMP parallelization for performance. The library handles NetCDF input, staggered grids, map projections, moving nests, and supports multiple plotting backends (Cartopy, Basemap, PyNGL). It is built with scikit-build-core and CMake, requires a Fortran compiler, and includes extensive unit tests comparing results against NCL reference data. The architecture follows a consistent decorator pattern for argument handling, iteration over leftmost dimensions, type casting, and metadata injection, making it a production-grade tool for atmospheric research and operational meteorology.