What data format is required for the Nixtla Model Selector?

Input data must be a CSV file containing three required columns: unique_id (series identifier), ds (datetimes), and y (target values).

Can this skill handle multiple time series simultaneously?

Yes, the skill is designed to handle multiple series identified by unique_id and even uses parallel processing (n_jobs=-1) when executing StatsForecast.

How does the skill decide between StatsForecast and TimeGPT?

It uses StatsForecast for datasets with missing values, seasonal patterns, high series counts, or short history (<30 points). It selects TimeGPT for long, clean, and non-seasonal data.

Do I need an API key to use this skill?

An API key is only required if the skill selects TimeGPT. In that case, you must have the NIXTLA_TIMEGPT_API_KEY environment variable configured.

What specific models does StatsForecast execute?

By default, the skill runs AutoETS and AutoARIMA within the StatsForecast framework to provide robust statistical benchmarks.

Nixtla Model Selector

Name: Nixtla Model Selector
Author: intent-solutions-io

byintent-solutions-io

データサイエンスとML

Automatically selects and executes the optimal forecasting model between StatsForecast and TimeGPT based on time series data characteristics.

概要

The Nixtla Model Selector is a specialized Claude Code skill designed to streamline time series forecasting by removing the guesswork from model selection. By analyzing data length, frequency, seasonality, and missing values, the skill intelligently routes datasets to either StatsForecast—best for short, seasonal, or sparse data—or TimeGPT, which is optimal for long-range complex patterns. This automation accelerates data science workflows, ensures the most accurate statistical or foundation models are used for specific datasets, and provides a clear decision rationale for every forecast generated.

主な機能

Support for large-scale batch forecasting across multiple series
Detailed selection reports explaining the logic behind chosen models
0 GitHub stars
Standardized long-format CSV output for seamless data integration
Intelligent routing between traditional statistical models and TimeGPT
Automated data characteristic analysis including seasonality and sparsity

ユースケース

Rapidly deploying production-ready forecasting services via FastAPI
Forecasting demand for retail products with varying historical data lengths
Automating model benchmarking for large-scale financial time series

概要

主な機能

Support for large-scale batch forecasting across multiple series
Detailed selection reports explaining the logic behind chosen models
0 GitHub stars
Standardized long-format CSV output for seamless data integration
Intelligent routing between traditional statistical models and TimeGPT
Automated data characteristic analysis including seasonality and sparsity

ユースケース

Rapidly deploying production-ready forecasting services via FastAPI
Forecasting demand for retail products with varying historical data lengths
Automating model benchmarking for large-scale financial time series