Prediction target and Features in Machine Learning
Updated
•1 min read
Prediction Target
It represents the output of an ML model.
For example, In a supervised learning algorithm, if you are developing a model for predicting house prices then your dataset will have a “Price” column, so this column will be your prediction target.• Only one column can be the prediction target.
Represented as “y”.
It can be accessed with a dot notation approach as follows.y = my_dataset.Price
Features
Features are nothing but the inputs that are provided to a model.
The feature can be any number of columns from the dataset, except for the prediction target.
Based on the selected input columns (features) the output will be predicted by the model.
Represented as “X” as follows.
features = [‘TotalRooms’, ‘Floors’, ‘Landsize’]X = my_dataset[melbourne_features]]
