ClimateNet

Model Introduction

ClimateNet is an expert-labeled extreme-weather dataset and pixel-level segmentation model for tropical cyclones and atmospheric rivers.

Paper: ClimateNet: an expert-labeled open dataset and deep learning architecture for enabling high-precision analyses of extreme weather
https://doi.org/10.5194/gmd-14-107-2021

Model Description

The model was proposed by teams from LBNL, UC Berkeley, ETH Zurich, NVIDIA, NCAR, and collaborators. It was trained with four-channel CAM5.1 fields and expert segmentation masks. DeepLabv3+ supports tropical-cyclone and atmospheric-river detection and conditional precipitation analysis.

Use Cases

Use Case Description
Extreme segmentation Identify background, TC, and AR pixels.
Climate scenarios Transfer segmentation to warming experiments.
Conditional precipitation Extract event-conditioned precipitation statistics.
ModelScope/OneCode execution Validate data, training, inference, segmentation metrics, and visualization.
Multi-GPU training Start multi-process training through torchrun.

Usage Instructions

hf download OneScience-Group/ClimateNet --local-dir ./ClimateNet
cd ClimateNet

Environment Dependencies

Hardware Requirements

  • A GPU or DCU is recommended.
  • A CPU can be used for connectivity validation with the default small-sample configuration.
  • DCU users should install DTK 25.04.2 or a compatible OneScience-recommended version first.

DCU Environment

# Activate DTK and Conda first
conda create -n onescience311 python=3.11 -y
conda activate onescience311
pip install onescience[earth-dcu] -i http://mirrors.onescience.ai:3141/pypi/simple/ --trusted-host mirrors.onescience.ai

GPU Environment

# Activate Conda first
conda create -n onescience311 python=3.11 -y libstdcxx-ng=12 libgcc-ng=12 gcc_linux-64=12 gxx_linux-64=12
conda activate onescience311
pip install onescience[earth-gpu] -i http://mirrors.onescience.ai:3141/pypi/simple/ --trusted-host mirrors.onescience.ai
python scripts/fake_data.py
python scripts/train.py
torchrun --standalone --nproc_per_node=2 scripts/train.py
python scripts/inference.py
python scripts/result.py

Training uses weighted cross-entropy. Inference returns finite class probabilities and evaluation reports per-class and mean IoU.

Trained Weights

No weights are bundled under weight/. The authors provide trained models and data at https://portal.nersc.gov/project/ClimateNet/.

Citation and License

This repository is an independent engineering reproduction of the public ClimateNet specifications.

The original paper is licensed under CC BY 4.0; official models, code, and data retain their respective terms.

Downloads last month
-
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support