Published December 21, 2023 | Version 1.0
Dataset Open

Model checkpoints for "SEEDS: Emulation of Weather Forecast Ensembles with Diffusion Models"

Authors/Creators

  • 1. Google Research

Description

Checkpoints for all SEEDS models in the paper https://arxiv.org/abs/2306.14066, including all SEEDS-GEE and SEEDS-GPP models in the main manuscript, and the additional models trained in the Supplemental Material.

Checkpoint naming convention:

  • gee_c2_s7: SEEDS-GEE trained conditioning on 2 seeds for 7-day lead time.
  • gpp_c2_s7_g3_r4: SEEDS-GPP trained conditioning on 2 seeds for 7-day leadtime, where the label mixture is 3 GEFS members and 4 ERA reanalyses.

Files

model_checkpoints.zip

Files (9.3 GB)

Name Size
md5:79a1970d515ce300aa2dbd0f403e3035
9.3 GB Preview Download

Additional details

Related works

Is published in
10.48550/arXiv.2306.14066 (DOI)