PCA (Principle Component Analysis)

Example Usage

1
2
3
4
5
6
7
8
9
10
11
12
13
from mltrain.unsupervised.PCA import PCA
import numpy as np

# Initialize the model
pca = PCA(n_components=2)

# Fit the model and transform the data and plot graph too.
transformed_X = pca.train_transform(X_train, plot_graph=True)

# Get the principal components
principal_components = pca.pc

Overview

The PCA class implements Principal Component Analysis (PCA), a technique for dimensionality reduction. PCA transforms data into a new coordinate system where the axes (principal components) are ordered by the amount of variance they capture from the data.

Hyperparameters

• n_components (int, default=2): The number of principal components to retain after dimensionality reduction.

Attributes

• pc (numpy.ndarray): The principal components (eigenvectors) after fitting the model.
• mean (numpy.ndarray): The mean of the features in the original data.

Methods

__init__(self, n_components=2)

Initializes the PCA model with the specified number of components.

• Args:
  • n_components (int): Number of principal components to retain.

train(self, X)

Fits the PCA model to the input data.

• Args:
  • X (numpy.ndarray): The input data to perform PCA on, with shape (n_samples, n_features).
• Returns:
  • numpy.ndarray: The principal components after fitting the model.
• Raises:
  • ValueError: If the number of components is greater than the number of features.

transform(self, X)

Applies the dimensionality reduction on the input data.

• Args:
  • X (numpy.ndarray): The input data to transform, with shape (n_samples, n_features).
• Returns:
  • numpy.ndarray: The data transformed into the principal component space.

train_transform(self, X, plot_graph=False)

Fits the PCA model and transforms the input data in one step. Optionally, plots the data in the reduced principal component space.

• Args:
  • X (numpy.ndarray): The input data to fit and transform, with shape (n_samples, n_features).
  • plot_graph (bool, optional): Whether to plot the transformed data. Only works for 1, 2, or 3 components.
• Returns:
  • numpy.ndarray: The data transformed into the principal component space.