Loading…
Glossary term
An organized collection of data used to train, tune or evaluate a model.
A dataset is an organized collection of data. For machine learning it is usually split into three parts: a training set to learn from, a validation set to tune choices, and a test set kept aside for a final honest score.
Well-known public datasets let researchers compare methods on equal footing.
Split a dataset into train and test
from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(
X, y, test_size=0.2, random_state=42
)The dataset defines the problem a model solves. A model is only as good, and as fair, as the data it was built on.