Multivariate-Linear-Regression

This is a code done from scratch.

Features

Generates n-dimensional data set

It generates a n-dimensional data set and then stores in a .csv file named data.csv.
To use this feature use the gen_data module datagen library. This module takes in 6 parameters,

• dimesion: Value of the dimension for the data set that needs to be genrated like 2, 3, 4, ... .
• x_lower_range: The lower range value of x for data generation.
• x_higher_range: The higher/upper range value of x for data generation.
• m_lower_range: The lower range value of m for data generation.
• x_higher_range: The higher range value of m for data generation.
• epoch: Number of data points.

Plots the models

After the model is trained 2D and 3D trained data models are plotted. Also Epoch VS Loss curve is plotted in the same window.
To use this feature use the showPlot module modelplot library. This module takes in 5 parameters,

• epochs: Epoch array for the Epoch vs Loss curve.
• lossed: Loss array for the Epoch vs Loss curve.
• dataSet: The whole data set that will be used to plot the model.
• mSet: The list of slopes of the trained model.
• c: The constant of the trained model.

Print tabular data after training

A comparission of y value of original dataset and y predicted value for particular data points. For this the test set will be used to find the y predicted.