skills.rest
SciMate-AI
SciMate-AIOFFICIAL
Explore all skills by this author →
GitHub

timesfm-forecasting

Forecast univariate time series with TimesFM and quantile intervals.

1|Updated Mar 11, 2026
Search Tags:#forecasting#preflight#timeseries#timesfm#covariates#quantiles
One-click install
npx skills add https://github.com/SciMate-AI/scicli --skill timesfm-forecasting-scimate-ai
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: timesfm-forecasting
Source: https://github.com/SciMate-AI/scicli/tree/main/internal/skills/bundled/claude-scientific-skills/skills/timesfm-forecasting
Command: npx skills add https://github.com/SciMate-AI/scicli --skill timesfm-forecasting-scimate-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires torch, timesfm, numpy, pandas, and includes scripts (resource) and references (resource) components.

What problem does it solve?

Timeseries forecasting for univariate data can be complex and resource-intensive; this skill provides a zero-shot TimesFM-based workflow with a mandatory preflight check to ensure RAM, GPU, and disk readiness before loading model weights, enabling reliable local inference.

Core Features & Use Cases

  • Zero-shot forecasting for any univariate time series without training.
  • Built-in preflight checks (RAM, GPU/VRAM, disk space, Python version) to prevent runtime crashes.
  • End-to-end guidance for common workflows including CSV/DataFrame/array inputs and generation of point forecasts with quantile intervals.

Quick Start

Run the preflight check and forecast a univariate time series with TimesFM to obtain a point forecast and prediction intervals.

Frequently Asked Questions about timesfm-forecasting

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I do zero-shot time-series forecasting on a CSV file without training a model?▼

You can perform zero-shot time-series forecasting directly on CSV inputs using TimesFM. The skill processes univariate data to generate point forecasts and quantile prediction intervals without requiring any model training.

What is zero-shot univariate forecasting and how do quantile prediction intervals work?▼

Zero-shot univariate forecasting predicts future values from a single data sequence without prior training. Quantile prediction intervals provide a range of expected outcomes, giving you both point forecasts and uncertainty bounds for your time series.

Do I need a GPU and specific RAM to run TimesFM for timeseries forecasting?▼

Running TimesFM requires sufficient RAM, GPU/VRAM, and disk space. The skill enforces a mandatory preflight check to verify your system resources and Python version before loading the model weights to prevent runtime crashes.

How do I use pandas DataFrames or numpy arrays for timeseries forecasting?▼

You can use pandas DataFrames or numpy arrays as direct inputs for univariate timeseries forecasting. The skill accepts these formats alongside CSV files to generate point forecasts and quantile intervals seamlessly.

Can I use torch and timesfm for forecasting without installing additional heavy dependencies?▼

You need torch, timesfm, numpy, and pandas installed to run forecasts. These dependencies support the underlying model operations and data handling required to generate predictions from your univariate time series.

Why does timeseries forecasting fail to load models due to memory or disk space limits?▼

Timeseries forecasting models fail to load when system resources are insufficient. The preflight check validates available RAM, GPU, and disk space beforehand to ensure reliable local inference and prevent runtime crashes during model loading.

Related Skills

Explore More Data & Analytics Skills→
vasilyu1983vasilyu1983
73

ai-ml-timeseries

Forecast time series data with LightGBM, Transformers, and SHAP.

Community
Advanced
RamanEbrahimiRamanEbrahimi
3

timesfm-forecasting

Generates zero-shot time-series forecasts and prediction intervals from CSV, DataFrame, or array inputs using Google's TimesFM foundation model and torch library.

Community
Advanced
AlterLab-IEUAlterLab-IEU
58

alterlab-timesfm

Forecast univariate time-series data with zero-shot TimesFM point and quantile forecasts.

Community
Advanced
Yezez9Yezez9

timesfm-forecasting

Forecast univariate time series with Google's TimesFM foundation model.

Community
Advanced