跳到正文
TakoAPI
返回技能列表

s2s-forecasting-expert

由 manmeet3591 提供

CLI Utilities
396下载量由 clawskills.sh 提供
0浏览量
0 个赞登录后即可为这个技能点赞
—暂无评分
你的评分

简介

End-to-end builder for AI-based Subseasonal-to-Seasonal (S2S) forecasting systems.

描述

该技能的功能

Generates runnable PyTorch code for Subseasonal-to-Seasonal (S2S) weather forecasting systems. Covers model architectures inspired by FuXi, FengWu, and AIFS, plus training loops, CRPS loss, and evaluation scripts. All code runs locally with no external API calls.

Provides domain-specific architecture patterns and loss functions for S2S forecasting that general code assistants rarely get right without extensive prompting.

何时使用

  • Building a FuXi-style transformer for 30-day Z500 forecasting
  • Implementing CRPS loss for probabilistic ensemble outputs
  • Scaffolding an ERA5 data preprocessing and training pipeline
  • Setting up distributed multi-GPU training for global grid models
  • Generating evaluation scripts comparing model skill against climatology

示例工作流

输入

User asks: Generate a FuXi-style transformer in PyTorch for 30-day Z500 forecasting with CRPS loss

智能体
  1. 1

    Designs a 3D spatiotemporal transformer architecture with global grid attention

  2. 2

    Generates multi-variable input pipeline handling Z500, T2M, and wind fields

  3. 3

    Implements lead-time conditioned decoder with ensemble output heads

  4. 4

    Creates CRPS loss function in both Gaussian and ensemble forms

  5. 5

    Outputs training loop with mixed precision and checkpoint saving

输出

Complete PyTorch module files for model, loss, and training loop ready to run locally

安装

clawhub install manmeet3591/s2s-forecasting-expert

运行 CLI 命令,或把这段提示词粘贴到 Claude Code、Cursor、Codex 或你在用的任意编程智能体里。

链接