Releases: wbtian/NoahPy
NoahPy v1.0.1
NoahPy v1.0.0.3
🔖 NoahPy v1.0.0.3 – Minor Bug Fixes & Test Script Added
✅ What's New
🐞 Fixed minor bugs in:
NoahPy.py
NoahPy_Module.py
🧪 Added a new test script test.py
This script demonstrates how to run NoahPy
When executed, it reads input from forcing.txt and outputs soil moisture and temperature data
NoahPy v1.0.0.2
📦 NoahPy v1.0.0.2 – Minor Update
🔧 What's New
🔄 DOI Updated: The project DOI has been updated to a unified All-version DOI to ensure consistent citation across versions.
📘 Added Chinese Technical Documentation: A Chinese-language user guide has been added to assist native speakers in understanding and using NoahPy.
📝 Improved README.md: The README has been updated with clearer usage instructions and links, improving usability and accessibility.
Feel free to report issues or contribute improvements — we welcome your feedback!
NoahPy v1.0.0 – Initial Release
📦 Release v1.0 — NoahPy: A Differentiable Noah LSM
Release Date: 2024-07-28
Version: v1.0.0
🧾 Overview
This is the first official release of NoahPy, a differentiable version of the Noah Land Surface Model (v3.4.1), re-implemented in PyTorch to enable gradient-based optimization and seamless integration with machine learning workflows.
The model preserves the core physical processes of Noah LSM while allowing automatic differentiation, making it suitable for data assimilation, parameter inversion, and physics-informed learning.
✅ Main Features
🌍 Based on Noah LSM v3.4.1 physics
🔁 Fully differentiable via PyTorch
🧠 Supports backpropagation and integration with neural networks
🚀 Getting Started
Install required packages (e.g., PyTorch ≥ 2.0)
Prepare input forcing data in standard format
Modify parameters in parameter_new/ if needed
Use NoahPy.py/NoahPy_Module.py as the main interface for forward/backward model computation