polynomial_batch_gd.c File Reference

Batch gradient descent linear regression example using polynomial features. More...

#include <stdio.h>
#include <dmlfw.h>
#include <stdlib.h>
#include <unistd.h>

Include dependency graph for polynomial_batch_gd.c:

#define	ACTUAL_TRAINING_DATASET   "trends_train.csv"
#define	FREQUENCY_OF_PRINTING_COST   300000
dmlfw_gradient_descent_options *	get_gradient_descent_options ()
	Creates gradient descent configuration for batch gradient descent.
FILE *	gnuplot
#define	LEARNING_RATE   0.09
void	load_dataset (dmlfw_mat_double **x, dmlfw_column_vec_double **y)
	Loads features and target from scaled polynomial dataset with bias column.
int	main ()
	Main program execution for polynomial batch gradient descent linear regression.
#define	MODEL_FILE_NAME   "example-4-model.csv"
#define	NUMBER_OF_ITERATIONS   300000000
int	on_iteration_complete (uint64_t iteration_number, void *y, void *predicted_y, void *model, double regularization_parameter)
	Callback invoked at each gradient descent iteration to log cost and update graphs.
void	print_error_and_exit ()
	Prints ml-framework error and exits.
#define	REGULARIZATION_PARAMETER   0.0
#define	SHOW_GRAPH   1
#define	TRAINING_DATASET   "scaled_poly_trends_train.csv"

Detailed Description

Batch gradient descent linear regression example using polynomial features.

Author

Mohammed Daniyal

Version

1.0

Date

2025-09-26

This example trains a linear regression model on the scaled polynomial trends training dataset. It tracks training cost with optional graphing, uses regularization, and saves the trained model. Includes custom progress callback with gnuplot visualization.

Usage: ./polynomial_batch_gd

Macro Definition Documentation

ACTUAL_TRAINING_DATASET

#define ACTUAL_TRAINING_DATASET   "trends_train.csv"

FREQUENCY_OF_PRINTING_COST

#define FREQUENCY_OF_PRINTING_COST   300000

LEARNING_RATE

#define LEARNING_RATE   0.09

MODEL_FILE_NAME

#define MODEL_FILE_NAME   "example-4-model.csv"

NUMBER_OF_ITERATIONS

#define NUMBER_OF_ITERATIONS   300000000

REGULARIZATION_PARAMETER

#define REGULARIZATION_PARAMETER   0.0

SHOW_GRAPH

#define SHOW_GRAPH   1

TRAINING_DATASET

#define TRAINING_DATASET   "scaled_poly_trends_train.csv"

Function Documentation

get_gradient_descent_options()

dmlfw_gradient_descent_options * get_gradient_descent_options

(

)

Creates gradient descent configuration for batch gradient descent.

Returns

Configured gradient descent options structure or NULL on error.

load_dataset()

void load_dataset	(	dmlfw_mat_double **	x,
		dmlfw_column_vec_double **	y
	)

Loads features and target from scaled polynomial dataset with bias column.

Parameters

[out]	x	Pointer to features matrix pointer.
[out]	y	Pointer to target column vector pointer.

main()

int main

(

)

Main program execution for polynomial batch gradient descent linear regression.

Returns

0 on success or an error code.

on_iteration_complete()

int on_iteration_complete	(	uint64_t	iteration_number,
		void *	y,
		void *	predicted_y,
		void *	model,
		double	regularization_parameter
	)

Callback invoked at each gradient descent iteration to log cost and update graphs.

Parameters

iteration_number	Current iteration index.
y	Actual target vector.
predicted_y	Predicted target vector.
model	Current model parameters vector.
regularization_parameter	Regularization coefficient.

Returns

0 to continue gradient descent, negative to abort.

print_error_and_exit()

void print_error_and_exit

(

)

Prints ml-framework error and exits.

Variable Documentation

gnuplot

FILE* gnuplot