Features (explanatory variables) are the input columns a model uses to predict; the target is the value to predict. In scikit-learn, features are gathered in a matrix X and the target in a vector y.
Example
X = df[['surface', 'nb_pieces']]
y = df['prix']