liveBook · Manning

Inference performance

As explained in Time Series Forecasting Using Foundation Models

Inference time is the duration required for a model to generate predictions from its input.

Inference performance measures how long a model takes to generate predictions. Faster inference can enable lower-latency decisions, while larger models may improve accuracy at substantially higher computational cost. section 6.1.3 section 10.2.9

Evaluate it alongside forecasting accuracy rather than treating speed as an isolated score. A practical comparison measures inference across cross-validation windows, then plots error against elapsed inference time. section 10.2 section 10.2.9

Daily website visitors

A capstone comparison measured seven-day forecasts over 20 windows. TimesFM and Chronos were slowest but most accurate, while TimeGPT and Moirai were among the fastest; the result was explicitly not a complete benchmark. section 10.2 section 10.2.9

How Time Series Forecasting Using Foundation Models covers it

How deep it goes
Measures elapsed prediction time across validation windows and compares it with error, but stops short of a complete benchmark.
Examples
Compares Seasonal Naive, SARIMA, TimeGPT, Chronos, Moirai, TimesFM, and Time-LLM on daily website-visitor forecasts.
What it assumes
Assumes familiarity with forecasting models, accuracy metrics, and cross-validation; teaches timing as an additional selection criterion.
What you can do after
Evaluate forecasting models by balancing prediction error against measured inference time and available computing resources.
How it connects
Connects model selection, forecasting accuracy, hardware resources, cross-validation, and latency.
Cover of Time Series Forecasting Using Foundation Models
liveBookmanning.com
Read Time Series Forecasting Using Foundation Models on liveBook Get the book at manning.com

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Time Series Forecasting Using Foundation Models. The section links point to the parts of the book it draws on.