This is the Linux app named Google DeepMind GraphCast and GenCast whose latest release can be downloaded as Version0.1.1sourcecode.tar.gz. It can be run online in the free hosting provider OnWorks for workstations.
Download and run online this app named Google DeepMind GraphCast and GenCast with OnWorks for free.
Urmați aceste instrucțiuni pentru a rula această aplicație:
- 1. Ați descărcat această aplicație pe computer.
- 2. Introduceți în managerul nostru de fișiere https://www.onworks.net/myfiles.php?username=XXXXX cu numele de utilizator pe care îl doriți.
- 3. Încărcați această aplicație într-un astfel de manager de fișiere.
- 4. Porniți emulatorul online OnWorks Linux sau Windows online sau emulatorul online MACOS de pe acest site web.
- 5. Din sistemul de operare OnWorks Linux pe care tocmai l-ați pornit, accesați managerul nostru de fișiere https://www.onworks.net/myfiles.php?username=XXXXX cu numele de utilizator dorit.
- 6. Descărcați aplicația, instalați-o și rulați-o.
CAPTURĂ DE ECRAN:
Google DeepMind GraphCast și GenCast
DESCRIERE:
GraphCast, developed by Google DeepMind, is a research-grade weather forecasting framework that employs graph neural networks (GNNs) to generate medium-range global weather predictions. The repository provides complete example code for running and training both GraphCast and GenCast, two models introduced in DeepMind’s research papers. GraphCast is designed to perform high-resolution atmospheric simulations using the ERA5 dataset from ECMWF, while GenCast extends the approach with diffusion-based ensemble forecasting for probabilistic weather prediction. Both models are built on JAX and integrate advanced neural architectures capable of learning from multi-scale geophysical data represented on icosahedral meshes. The package includes pretrained model weights, normalization statistics, and demonstration notebooks that allow users to replicate and fine-tune weather forecasting experiments in Colab or on Google Cloud TPUs and GPUs.
Categorii
- Implements GraphCast and GenCast architectures for data-driven weather forecasting
- Pretrained model weights and normalization data available via Google Cloud Bucket
- JAX-based differentiable simulation framework using graph neural networks
- Colab-ready demonstration notebooks for quick experimentation and learning
- Compatible with ERA5 and HRES datasets for historical and operational fine-tuning
- Supports execution on TPUs and GPUs for scalable model training and inference
Limbaj de programare
Piton
Categorii
This is an application that can also be fetched from https://sourceforge.net/projects/g-deepmind-graph-gen.mirror/. It has been hosted in OnWorks in order to be run online in an easiest way from one of our free Operative Systems.