Neural Networks

The MLPRegressor (Multi-layer Perceptron Regressor) is a neural network model that learns a non-linear function for regression by training on datasets using backpropagation. It consists of at least three layers and utilizes activation functions like ReLU or Logistic to model complex dependencies. Unlike linear regressors, it can capture intricate patterns, though it requires careful tuning of hyperparameters and feature scaling to perform optimally.

Pros neural networks:

• Can understand very very complex relationships.
• Semi-resilient to over fitting.

Cons neural networks:

• Very very slow to train
• requires tuning of Model Hyper Parameters to reach full potential.

Regressors

Model Name	Summary	Link to sklearn-documentation
MLPRegressor	A supervised learning algorithm that trains using backpropagation over multiple layers of nodes (neurons) to predict continuous outputs through non-linear transformations.	Documentation

Classifiers

Model Name	Summary	Link to sklearn-documentation
MLPClassifier	A supervised learning algorithm that trains using backpropagation over multiple layers of nodes (neurons) to predict continuous outputs through non-linear transformations.	Documentation