Forecasting
Explore how to use ML.NET for forecasting by building models that predict future values based on historical time series data. Understand the importance of features like time columns and forecasting horizons, and learn how to utilize the ForecastBySsa trainer to handle complex data patterns and generate predictions for various applications.
We'll cover the following...
Forecasting involves predicting future values or trends based on historical data patterns. This task is commonly used in various domains, such as sales forecasting, stock market prediction, demand forecasting, weather forecasting, and more. The following figure summarizes the forecasting process:
Once the model is trained, it can be used for forecasting future values. The model takes in new input data representing future time points and generates predictions based on the learned patterns. The ultimate goal of a forecasting ML task is to generate reliable predictions that support decision-making and planning in various domains.
Building a forecasting model
In the following playground, we have a dataset that represents electricity prices over time. It has some additional information, such as the day of the week, whether ...