liveBook · Manning

Neuralforecast

As explained in Time Series Forecasting Using Foundation Models

NeuralForecast is an object in the neuralforecast library that manages data windows, batching, and model training.

NeuralForecast is a Python library object that prepares time-series windows, batches data, trains forecasting models, and generates predictions through a simple interface. section 2.4 section 9.2.1

Use NeuralForecast to:

  • Initialize models with a forecast horizon, input size, frequency, batch settings, and maximum training steps. section 2.4 section 9.2.1
  • Fit models using gradient-update steps rather than dataset epochs. section 2.4
  • Load pretrained models and produce zero-shot forecasts for new series. section 2.5
Listing 9.3 Initializing the NeuralForecast object
nf = NeuralForecast(
    models=[timellm],
    freq='W'
)

Shows the minimal initialization pattern: provide models and the dataset frequency before training or forecasting. section 9.2.1

How Time Series Forecasting Using Foundation Models covers it

How deep it goes
Explains initialization, training steps, frequency settings, and zero-shot prediction, but defers implementation internals.
Examples
Demonstrates monthly M3 pretraining, weekly store-sales forecasting, and Australian antidiabetic-prescription prediction.
What it assumes
Assumes Python forecasting data with unique_id, ds, and y columns while teaching the required interface through examples.
What you can do after
Operate/configure forecasting workflows by initializing models, fitting them, saving or loading them, and requesting predictions.
How it connects
Connects N-BEATS, Time-LLM, transfer learning, zero-shot forecasting, batching, and time-series windows.
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.