LuoJianing 4902dfe31b | 1 month ago | |
---|---|---|
.. | ||
applications | 1 month ago | |
cmake | 3 months ago | |
docs | 3 months ago | |
mindearth | 1 month ago | |
CMakeLists.txt | 3 months ago | |
CONTRIBUTION_CN.md | 3 months ago | |
LICENSE.txt | 3 months ago | |
NOTICE.txt | 3 months ago | |
README.md | 3 months ago | |
README_CN.md | 3 months ago | |
RELEASE.md | 2 months ago | |
RELEASE_CN.md | 2 months ago | |
build.sh | 3 months ago | |
requirements.txt | 3 months ago | |
setup.py | 3 months ago | |
version.txt | 2 months ago |
ENGLISH | 简体中文
Weather phenomena are closely related to human production and life, social economy, military activities and other aspects. Accurate weather forecasts can mitigate the impact of disaster weather events, avoid economic losses, and generate continuous fiscal revenue, such as energy, agriculture, transportation and entertainment industries. At present, the weather forecast mainly adopts numerical weather prediction models, which processes the observation data collected by meteorological satellites, observation stations and radars, solves the atmospheric dynamic equations describing weather evolution, and then provides weather and climate prediction information. The prediction process of numerical prediction model involves a lot of computation, which consumes a long time and a large amount of computation resources. Compared with the numerical prediction model, the data-driven deep learning model can effectively reduce the computational cost by several orders of magnitude.
MindEarth is an earth science suite developed based on MindSpore. It supports AI meteorological prediction of short-term, medium-term, and long-term weather and catastrophic weather such as precipitation and typhoon. The aim is to provide efficient and easy-to-use AI meteorological prediction software for industrial scientific research engineers, college teachers and students.
2023.02.06
MindSpore Helps Ocean Terrain Overscore: Huang Xiaomeng Team of Tsinghua University Releases Global 3 Arc Second (90 m) Sea and Land DEM Data Products.Page.Case | Dataset | Network | GPU | NPU |
---|---|---|---|---|
DGMs | Radar dataset | GAN、ConvGRU | ✔️ | ✔️ |
Case | Dataset | Network | GPU | NPU |
---|---|---|---|---|
FourCastNet | ERA5 Reanalysis Dataset | AFNO | ✔️ | ✔️ |
ViT-KNO | ERA5 Reanalysis Dataset | ViT | ✔️ | ✔️ |
GraphCast | ERA5 Reanalysis Dataset | GNN | ✔️ | ✔️ |
Case | Dataset | Network | GPU | NPU |
---|---|---|---|---|
DEM Super-resolution | NASADEM、GEBCO_2021 | SRGAN | ✔️ | ✔️ |
Because MindEarth is dependent on MindSpore, please click MindSpore Download Page according to the corresponding relationship indicated in the following table. Download and install the corresponding whl package.
MindEarth | Branch | MindSpore | Python |
---|---|---|---|
master | master | \ | >=3.7 |
0.1.0 | r0.5 | >=1.8.1 | >=3.7 |
pip install -r requirements.txt
Hardware | OS | Status |
---|---|---|
Ascend 910 | Ubuntu-x86 | ✔️ |
Ubuntu-aarch64 | ✔️ | |
EulerOS-aarch64 | ✔️ | |
CentOS-x86 | ✔️ | |
CentOS-aarch64 | ✔️ | |
GPU CUDA 11.1 | Ubuntu-x86 | ✔️ |
# gpu and ascend are supported
export DEVICE_NAME=gpu
pip install mindearth_${DEVICE_NAME}
git clone https://gitee.com/mindspore/mindscience.git
cd {PATH}/mindscience/MindEarth
bash build.sh -e ascend -j8
export CUDA_PATH={your_cuda_path}
bash build.sh -e gpu -j8
cd {PATH}/mindscience/MindEarth/output
pip install mindearth_*.whl
Thanks goes to these wonderful people 🧑🤝🧑:
yufan, wangzidong, liuhongsheng, zhouhongye, liulei, libokai, chengqiang, dongyonghan, zhouchuansai
MindScience is scientific computing kits for various industries based on the converged MindSpore framework.
Jupyter Notebook Python Unity3D Asset Pickle nesC other
Dear OpenI User
Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.
For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》