alterlab-timesfm

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

58|9|Updated Mar 16, 2026

Search Tags:#forecasting #time-series #preflight #system-check #timesfm #univariate

Ready to run

npx skills add https://github.com/AlterLab-IEU/AlterLab-Academic-Skills --skill alterlab-timesfm

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: alterlab-timesfm
Source: https://github.com/AlterLab-IEU/AlterLab-Academic-Skills/tree/main/skills/data-science/alterlab-timesfm
Command: npx skills add https://github.com/AlterLab-IEU/AlterLab-Academic-Skills --skill alterlab-timesfm

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill enables zero-training, zero-shot forecasting for univariate time series using Google's TimesFM, delivering both point forecasts and calibrated quantile intervals to support proactive decision-making.

Core Features & Use Cases

Zero-shot forecasting for any univariate time series (sales, sensors, energy, vitals, weather) without training a custom model.
Probabilistic outputs with calibrated prediction intervals to quantify uncertainty.
Preflight system checker to verify RAM, GPU availability, and disk space before loading the model.

Quick Start

Run the preflight check and then load TimesFM to forecast your time series with a prepared input.

Frequently Asked Questions about alterlab-timesfm

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

FAQPage Schema

How do I run zero-shot time-series forecasting without training a model?▼

Zero-shot time-series forecasting uses Google's TimesFM to predict univariate data like sales or sensor readings without custom training. This Skill loads the pre-trained model to deliver point forecasts and quantile intervals directly from your input arrays or DataFrames.

Can I forecast time-series data from a pandas DataFrame or CSV file?▼

Yes, time-series forecasting with this Skill accepts CSV, pandas DataFrame, or array inputs. It processes your univariate series to generate point forecasts and calibrated quantile intervals using the TimesFM model.

Do I need a GPU to run TimesFM for univariate forecasting?▼

A GPU is not strictly required but is checked during the preflight system check. The Skill verifies available RAM, GPU, and disk space before loading TimesFM to ensure your environment can handle the forecasting workload.

What is the best way to get prediction intervals for univariate time series?▼

The best way to get prediction intervals is using a zero-shot model like TimesFM. It provides calibrated quantile intervals alongside point forecasts, allowing you to quantify uncertainty without training a custom model.

Does TimesFM support batch forecasting for multiple sensor or sales series?▼

Yes, TimesFM supports batch forecasting for multiple univariate series. The Skill compiles batch forecasting capabilities using ForecastConfig to process several time series like sensor readings or sales data simultaneously.

Why does TimesFM fail to load during time-series forecasting?▼

TimesFM might fail to load if preflight system checks detect insufficient RAM, GPU unavailability, or low disk space. The Skill runs these checks before model loading to prevent crashes during the forecasting process.