import math
import random
import _thread
from GradientDescent import gradientDescent, getTrainingDataSet, readFile
def f(x, theta):
return theta[0] * x * x * x * x + theta[1] * x * x * x + theta[2] * x * x + theta[3] * x + theta[4]
arr = readFile()
for i in arr:
plt.scatter(i[0], i[1])
features, target = getTrainingDataSet(arr)
theta = [random.uniform(-1.0, 1.0), random.uniform(-1.0, 1.0), random.uniform(-1.0, 1.0), random.uniform(-1.0, 1.0), random.uniform(-1.0, 1.0)]
newtheta = gradientDescent(theta, features, target)
print('theta:\n')
for i in newtheta:
print(i, '\n')
x = np.arange(-5.0, 5.0, 0.01)
y = f(x, theta)
z = f(x, [0.0, 3.0, -4.0, 4.0, 2.0])
plt.plot(x, z, 'r')
plt.plot(x, y, 'b')
plt.xlabel('x')
plt.ylabel('y')
plt.show()
# Defining parameters for gradient descent
initalWeights = np.random.random([1,X.shape[1]])*0
maxIter = 1000
learningRate = 0.01
# Training a Least Squares Regression
weights = initalWeights
cost = []
for i in range(maxIter):
yp = np.matmul(X_train,weights.T).ravel()
J = computeL2Cost(Y_train, yp)
G = computeGradient(X_train, Y_train, yp)
weights = gradientDescent(weights, G, learningRate)
if i%10 ==0:
print("Cost of the model is {}".format(J))
cost.append(J)
print("Weights after the training are : {}".format(weights))
# Plotting the training loss curve
plt.plot(range(0,len(cost)),cost)
plt.title('Cost per iterations')
plt.show()
# Computing the Residuals of the Training data
# Defining parameters for gradient descent
initalWeights = np.random.random([1, X.shape[1]])
maxIter = 1000
learningRate = 0.3
# Training a Logistic Regression
weights = initalWeights
cost = []
for i in range(maxIter):
yp = logistic(weights, X_train)
J = computeEntropyCost(Y_train, yp)
G = computeGradient(X_train, Y_train, yp)
weights = gradientDescent(weights, G, learningRate)
if i % 10 == 0:
print("Cost of the model is {}".format(J))
cost.append(J)
print("Weights {} after the training are".format(weights))
# Plotting the training loss curve
plt.plot(range(0, len(cost)), cost)
plt.title('Cost per iterations')
plt.show()
# Prediction using the model
yp = logistic(weights, X_test)
yp = ((yp >= 0.5) * 1.0).ravel()
for row in r:
[x, y, z] = int(row[0]), int(row[1]), int(row[2])
a.append([x, y])
b.append(z)
'''Initializing values'''
X = matrix(a) # creating the x-data matrix
y = transpose(matrix(b)) # creating the y-data matrix
theta = zeros((3, 1)) # initializing theta value
alpha = 0.1 # initializing learning rate
num_iter = 400 # initializing no. of iterations
print('Area No. of rooms Cost')
for i, j in zip(X, y):
print('%d %d %d' % (i[0, 0], i[0, 1], j[0, 0]))
[X, mu, sigma] = featureNormalize(X)
X = insert(X, 0, 1, axis=1)
[theta, J_history] = gradientDescent(X, y, theta, alpha, num_iter)
'''Trying to predict the cost of house with sample area 1650 sq.m and 3 rooms'''
to_find = [1650, 3]
normalized = (to_find - mu) / sigma
normalized = insert(normalized, 0, 1, axis=1)
pcost = normalized * theta
print('#########Sample Prediction#########')
print('Price of house with area %d sq.m and %d rooms is ' %
(to_find[0], to_find[1])),
print('$%0.2f' % pcost[0, 0])
print('###################################')
'''Predicting house prices using custom inputs'''
to_find = []
x = int(input('Enter area of house: '))
y = int(input('Enter number of rooms: '))
to_find = [x, y]
