Skip to content
TakoAPI
Back to skills

s2s-forecasting-expert

by manmeet3591

CLI Utilities
396Downloads reported by clawskills.sh
0Views
0 likesSign in to like this skill
—Not yet rated
Your rating

Brief

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

Description

What This Skill Does

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.

When to Use It

  • 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

Example Workflow

INPUT

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

AGENT
  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

OUTPUT

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

Installation

clawhub install manmeet3591/s2s-forecasting-expert

Run the CLI command, or paste the prompt into Claude Code, Cursor, Codex or whichever coding agent you use.

Links