Deep Learning with PyTorch

Feature engineering

Feature engineering is the process of using domain knowledge about a particular problem to create new variables or features that can be passed to the model. To understand better, let's look at a sales prediction problem. Say we have information about promotion dates, holidays, competitor's start date, distance from competitor, and sales for a particular day. In the real world, there could be hundreds of features that may be useful in predicting the prices of stores. There could be certain information that could be important in predicting the sales. Some of the important features or derived values are:

• Days until the next promotion
• Days left before the next holiday
• Number of days the competitor's business has been open ...