本文へスキップ
TakoAPI
スキル一覧に戻る

s2s-forecasting-expert

manmeet3591 作

CLI Utilities
396ダウンロード数はclawskills.shが公開している値です
0閲覧数
0件のいいねサインインするとこのスキルにいいねできます
—まだ評価がありません
あなたの評価

概要

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

説明

このスキルでできること

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.

使いどころ

  • 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

ワークフロー例

入力

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

エージェント
  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

出力

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

インストール

clawhub install manmeet3591/s2s-forecasting-expert

CLIコマンドを実行するか、Claude Code、Cursor、Codexなどお使いのコーディングエージェントにプロンプトを貼り付けてください。

リンク