Vectorization Of Gradient Descent

In Machine Learning, Regression problems can be solved in the following ways:

1. Using Optimization Algorithms - Gradient Descent

Batch Gradient Descent.
Stochastic Gradient Descent.
Mini-Batch Gradient Descent
Other Advanced Optimization Algorithms like ( Conjugate Descent ... )

2. Using the Normal Equation :

Using the concept of Linear Algebra.

Let's consider the case for Batch Gradient Descent for Univariate Linear Regression Problem.

The cost function for this Regression Problem is :

J(\Theta)=(1/2m)*\sum_{i=1}^m(h_{\theta}(x^i)-y^i)^2

Goal:

minimize_{\ \theta_{o},\theta_{1}}\ \ J({\theta})

In order to solve this problem, we can either go for a Vectorized approach ( Using the concept of Linear Algebra ) or unvectorized approach (Using for-loop).

1. Unvectorized Approach:

Here in order to solve the below mentioned mathematical expressions, We use for loop.

The above mathematical expression is a part of Cost Function.

\sum_{i=1}^m(h_{\theta}(x^i)-y^i)^2

The above Mathematical Expression is the hypothesis.

h_{\theta}=\theta_{0}x_{0}+\theta_{1}x_{1}+\theta_{2}x_{2}+... +\theta_{n}x_{n}\\ where,\\ h_{\theta}=hypothesis.\\

Code: Python Implementation of Unvectorzed Grad

python3

# Import required modules.
from sklearn.datasets import make_regression
import matplotlib.pyplot as plt
import numpy as np
import time
 
# Create and plot the data set.
x, y = make_regression(n_samples = 100, n_features = 1,
                       n_informative = 1, noise = 10, random_state = 42)

plt.scatter(x, y, c = 'red')
plt.xlabel('Feature')
plt.ylabel('Target_Variable')
plt.title('Training Data')
plt.show()

# Convert y from 1d to 2d array.
y = y.reshape(100, 1)
 
# Number of Iterations for Gradient Descent
num_iter = 1000
 
# Learning Rate
alpha = 0.01
 
# Number of Training samples.
m = len(x)
 
# Initializing Theta.
theta = np.zeros((2, 1),dtype = float)
 
# Variables
t0 = t1 = 0
Grad0 = Grad1 = 0

# Batch Gradient Descent.
start_time = time.time()
 
for i in range(num_iter):
    # To find Gradient 0.
    for j in range(m):
        Grad0 = Grad0 + (theta[0] + theta[1] * x[j]) - (y[j])
    
    # To find Gradient 1.
    for k in range(m):
        Grad1 = Grad1 + ((theta[0] + theta[1] * x[k]) - (y[k])) * x[k]
    t0 = theta[0] - (alpha * (1/m) * Grad0)
    t1 = theta[1] - (alpha * (1/m) * Grad1)
    theta[0] = t0
    theta[1] = t1
    Grad0 = Grad1 = 0
     
# Print the model parameters.    
print('model parameters:',theta,sep = '\n')
 
# Print Time Take for Gradient Descent to Run.
print('Time Taken For Gradient Descent in Sec:',time.time()- start_time)

# Prediction on the same training set.
h = []
for i in range(m):
    h.append(theta[0] + theta[1] * x[i])
     
# Plot the output.
plt.plot(x,h)
plt.scatter(x,y,c = 'red')
plt.xlabel('Feature')
plt.ylabel('Target_Variable')
plt.title('Output')

Output:

model parameters:
[[ 1.15857049]
 [44.42210912]]
 
Time Taken For Gradient Descent in Sec: 2.482538938522339

2. Vectorized Approach:

Here in order to solve the below mentioned mathematical expressions, We use Matrix and Vectors (Linear Algebra).

The above mathematical expression is a part of Cost Function.

\sum_{i=1}^m(h_{\theta}(x^i)-y^i)^2

The above Mathematical Expression is the hypothesis.

h_{\theta}=\theta^T.X\\ where,\\ h_{\theta}=hypothesis.\\ \theta= \begin{bmatrix} \theta_{0} \\ \theta_{1}\\ \theta_{2}\\ \theta_{3}\\ .\\ .\\ \theta_{n}\\ \end{bmatrix} X= \begin{bmatrix} {x_{0}} \\ {x_{1}}\\ {x_{2}}\\ {x_{3}}\\ .\\ .\\ {x_{n}}\\ \end{bmatrix}\\

Batch Gradient Descent :

Loop\ until\ converge\{\\ \ \theta_{j}:=\theta_{j}-(1/m)*(\alpha)*\frac{\partial J(\theta)}{\partial \theta_{j}} \\ \}\\ Let, \ Gradients=\frac{\partial J(\theta)}{\partial \theta_{j}}

Concept To Find Gradients Using Matrix Operations:

X\_New= \begin{bmatrix} {x_{0}^1} & {x_{1}^1} \\ {x_{0}^2} & {x_{1}^2}\\ {x_{0}^3} & {x_{1}^3}\\ {x_{0}^4} & {x_{1}^4}\\ . & .\\ . & .\\ . & .\\ {x_{0}^m} & {x_{1}^m} \end{bmatrix}_{m X 2} \theta= \begin{bmatrix} \theta_{0} \\ \theta_{1}\\ \end{bmatrix}_{2X1}\\ where,\\ \x_{0}^i=1\\

H(\theta)=X\_New\ .\ \theta\\ H(\theta)= \begin{bmatrix} {\Theta_{0}}{x_{0}^1}+{\Theta_{1}}{x_{1}^1} \\ {\Theta_{0}}{x_{0}^2}+{\Theta_{1}}{x_{1}^2}\\ {\Theta_{0}}{x_{0}^3}+{\Theta_{1}}{x_{1}^3}\\ {\Theta_{0}}{x_{0}^4}+{\Theta_{1}}{x_{1}^4}\\ .\\ .\\ . \\ {\Theta_{0}}{x_{0}^m}+{\Theta_{1}}{x_{1}^m} \end{bmatrix}_{mX1} And\ \ \ Y= \begin{bmatrix} {y^1}\\ {y^2}\\ {y^3}\\ .\\ .\\ .\\ {y^m} \end{bmatrix}_{mX1}\\

H(\theta)-Y= \begin{bmatrix} {\Theta_{0}}{x_{0}^1}+{\Theta_{1}}{x_{1}^1} -y^1\\ {\Theta_{0}}{x_{0}^2}+{\Theta_{1}}{x_{1}^2}-y^2\\ {\Theta_{0}}{x_{0}^3}+{\Theta_{1}}{x_{1}^3}-y^3\\ {\Theta_{0}}{x_{0}^4}+{\Theta_{1}}{x_{1}^4}-y^4\\ .\\ .\\ . \\ {\Theta_{0}}{x_{0}^m}+{\Theta_{1}}{x_{1}^m}-y^m \end{bmatrix}_{mX1} \\

X\_New^T= \begin{bmatrix} {x_{0}^1\ x_{0}^2\ x_{0}^3\ .\ .\ .\ x_{0}^m}\\ {x_{1}^1\ x_{1}^2\ x_{1}^3\ .\ .\ .\ x_{1}^m} \end{bmatrix}_{2Xm} \\

Gradients=X\_New\ . \ (H(\theta)-Y)\\ = \begin{bmatrix} {x_{0}^1(\Theta x_{0}^1+\Theta x_{1}^1-y^1)\ + \ x_{0}^2(\Theta x_{0}^2+\Theta x_{1}^2-y^2)\ + \ x_{0}^3(\Theta x_{0}^3+\Theta x_{1}^3-y^3)\ + . . .}\\ {x_{1}^1(\Theta x_{0}^1+\Theta x_{1}^1-y^1)\ + \ x_{1}^2(\Theta x_{0}^2+\Theta x_{1}^2-y^2)\ + \ x_{1}^3(\Theta x_{0}^3+\Theta x_{1}^3-y^3)\ + . . .}\\ \end{bmatrix}_{2X1}\\

Finally\ we\ can \ say,\\ \ \ \ Gradients=\frac{\partial J(\theta)}{\partial \theta_{j}}=X\_New^T.(X\_New.\theta-Y)

Code: Python implementation of vectorized Gradient Descent approach