Training set

What is a training set?

Training set (training dataset) is a set of examples used to initially train an ML model. It's used to fit the model's parameters, for instance, the weights of connections between neurons in a neural network.

A training set can contain labeled and/or unlabeled data, depending on a machine learning paradigm used for the model training. Supervised learning relies on labeled data, and unsupervised learning utilizes unlabeled data.

Back to all