How to use GaussianNB (sklearn)

TOP MENU

Python

library
・pip

・MeCab

Common
・class

・pickle

・read/readline

numpy
・asfarray

・digitize

・expit

・linalg.solve

・meshgrid

・mgrid

・ndmin

・pad

・poly1d

・polyfit

・prod

・shape

matplotlib
・figure

・pcolormesh

・scatter

pytorch
・BCELoss, MSELoss

・device

・Embedding

・TensorDataset, Dataloader

・RNN, LSTM

scikit-learn
・SVC

・GaussianNB

scipy
・interpolate

tkinter
・postscript

・image display

・frame, grid

・Crop Image

other
・linear interpolation

・Hysteresis switch

・Square/Triangle wave

OpenAI gym
・CartPole-v0

By purpose
・1 of K Coding

Release date:2024/7/30　　　　　　　　　

・In Japanese

■Description of the GaussianNB function

This is a naive Bayes classifier whose probability distribution follows a Gaussian distribution. Gaussian naive Bayes finds the Gaussian distribution of data belonging to the same label, and determines which distribution new data is closer to.

■A concrete example of the GaussianNB function

In the graph below, the label values of the test data are classified from the feature data.

The program is as follows. To install various libraries such as scikit-learn, click here.

from sklearn.naive_bayes import GaussianNB
import numpy as np
import matplotlib.pyplot as plt

X = np.array([[5,10],[8,50],[10,20]])  # Feature data
Y = np.array([1, 2, 3])    # Feature data labels
t = np.array([[9,20]])    # test data

clf = GaussianNB()    # Definition of Gaussian Naive Bayes
clf.fit(X, Y)    # Fitting to feature data
print(clf.predict(t))    # Classify the test data

plt.plot([5,8,10],[10,50,20],'o')    # Plotting feature data
plt.plot([9],[20],'^')    # Plotting the test data
plt.grid(True)
plt.show()

The result should have returned a value of 3. This is certainly close to the value of 3 in the feature data, so the result seems reasonable.

List of related articles