# -*- coding: utf-8 -*-

"""Untitled5.ipynb

Automatically generated by Colaboratory.

Original file is located at

https://colab.research.google.com/drive/1P3KHjkaBoBiwK2K023s5f6Z45CrUDoWy

# PAVAN KUMAR REDDYMASU

#11525000
"""

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from sklearn.linear_model import LinearRegression, Ridge, Lasso
from sklearn.model_selection import train_test_split, cross_val_score
#from statistics import meanimport pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from sklearn.linear_model import LinearRegression, Ridge, Lasso
from sklearn.model_selection import train_test_split, cross_val_score
from statistics import mean

X=np.arange(9).reshape(9,1)
Y=np.array([2.5,4.2,2,1,4.4,0.9,1.7,5.1,12])

"""#1.1 plot the data"""

plt.title("Matplotlib plot")
plt.xlabel("x axis caption")
plt.ylabel("y axis caption")
plt.plot(X,Y)
plt.show()

"""#1.2 Train linear regressor y=ax+b"""

model = LinearRegression().fit(X, Y)
r_sq = model.score(X, Y)
print('coefficient of determination:', r_sq)

"""#1.3 Plot the coefficient and plot the model curve from 0 to 10 at
x=np.arrange(0,10,0.01)

"""

plt.scatter(X,Y,c='steelblue', edgecolor='white', s=70)

plt.xlabel('X values')
plt.ylabel('Y values')
plt.title('scatter plot of x vs y')
plt.show()

# Print the Intercept:

print('intercept:', model.intercept_)
# Print the Slope:
print('slope:', model.coef_)
# Predict a Response and print it:
y_pred = model.predict(X)
print('Predicted response:', y_pred, sep='\n')
"""#1.4 Use the model to predict x_new and plot x_new and its corresponding
predictions. keep the training data scatter as background in the same plot.

"""

x_new=np.arange(0,8,0.01).reshape(800,1)

y_new = model.predict(x_new)
print (y_new)

"""#1.5 Use the normal equation to solve the model paramters. Print the solution.

# Convert target variable array from 1d to 2d.

"""

x_new = np.array([np.ones(len(X)), X.flatten()]).T

x_new

"""# Using Normal Equation.

"""

theta_best_values = np.linalg.inv(x_new.T.dot(x_new)).dot(x_new.T).dot(Y)

# Display best values obtained.

print ("Display best values obtained")
print(theta_best_values)
# Plot the output.
plt.scatter(X,Y,s=30,marker='o')
plt.plot(X,y_pred,c='red')
plt.plot()
plt.xlabel("Feature")
plt.ylabel("Target_Variable")
plt.title('Linear Regression')
plt.show()

"""#1.6 Generate new data matrix with higher polynomial orders.

"""

from sklearn.preprocessing import PolynomialFeatures

#poly order2
poly = PolynomialFeatures(degree = 2)
X_poly = poly.fit_transform(X)
poly.fit(X_poly, Y)
lin2 = LinearRegression(fit_intercept=True)
lin2.fit(X_poly, Y)

w_0 = lin2.intercept_
w_1 = lin2.coef_[1]
w_2 = lin2.coef_[2]

print (lin2.intercept_)
print (lin2.coef_)

x_new=np.arange(0,8,0.01).reshape(800,1)
x_new_poly=poly.fit_transform(x_new)
prediction = lin2.predict(x_new_poly)

plt.scatter(x_new, prediction)
plt.show()

#poly order8
poly = PolynomialFeatures(degree = 8)

X_poly = poly.fit_transform(X)
lin = LinearRegression()
poly.fit(X_poly, Y)
lin3 = LinearRegression(fit_intercept=True)

lin3.fit(X_poly, Y)
w_0 = lin3.intercept_
w_1 = lin3.coef_[1]
w_2 = lin3.coef_[2]

print(lin3.intercept_)
print(lin3.coef_)

x_new=np.arange(0,8,0.01).reshape(800,1)
x_new_poly=poly.fit_transform(x_new)
prediction = lin3.predict(x_new_poly)

plt.scatter(x_new, prediction)
plt.show()

"""#2.1 Load iris data

"""

from sklearn.datasets import make_moons as mm

from sklearn.metrics import accuracy_score
from sklearn.model_selection import train_test_split
from sklearn.datasets import load_iris
iris= load_iris()
type(iris)
print (iris.data)
iris.data.shape
print (iris.feature_names)
from sklearn.datasets import make_moons as mm

X, y = mm(n_samples=200, noise=0.1, random_state=0)

plt.scatter(X[:,0], X[:,1], c=y)
plt.show()

"""#2.2 Split the data and report test accuracy

"""

# Split the data

X_train, X_test, Y_train, Y_test = train_test_split(iris['data'], iris['target'],
random_state=0)

# Prepare the classifier

from sklearn.linear_model import LogisticRegression
logreg = LogisticRegression()
# Fit the model
logreg.fit(X_train, Y_train)

logreg = LogisticRegression(penalty='l2', C=1.0)

logreg.fit(X_train, Y_train)
y_pred = logreg.predict(X_test)

print("Accuracy:", accuracy_score(Y_test, y_pred))

"""#2.3 Report L2 regularized test accuracy

# Evaluate the model
"""

print("Training scores: {:.2f}".format(logreg.score(X_train, Y_train)))

print("Test scores: {:.2f}".format(logreg.score(X_test,Y_test)))

print ("\n\n\n\n\n")

def find_theta(X, y):

m = X.shape[0] # Number of training examples.

# Appending a cloumn of ones in X to add the bias term.
X = np.append(X, np.ones((m,1)), axis=1)
# reshaping y to (m,1)
y = y.reshape(m,1)

# The Normal Equation

theta = np.dot(np.linalg.inv(np.dot(X.T, X)), np.dot(X.T, y))

return theta
def predict(X):

# Appending a cloumn of ones in X to add the bias term.

X = np.append(X, np.ones((X.shape[0],1)), axis=1)

# preds is y_hat which is the dot product of X and theta.

preds = np.dot(X, theta)

return preds

"""# Getting the Value of theta using the find_theta function.

"""

theta = find_theta(X, Y)
print ("The best values presented is as shown below: ", theta)

"""# Getting the predictions on X using the predict function.

"""

preds = predict(X)
# Plotting the predictions.
fig = plt.figure(figsize=(8,6))
plt.plot(X, Y, 'b.')
plt.plot(X, preds, 'c-')
plt.xlabel('X - Input')
plt.ylabel('y - target / true')
plt.show()

"""#2.4 Convert data to polynomial order 8. Report L2 regularized test accuracy and
Report none regularized test accuracy

"""

logreg = LogisticRegression(C = .01).fit(X_train, Y_train)

# Evaluate the model
print("Training scores: {:.2f}".format(logreg.score(X_train, Y_train)))
print("Test scores: {:.2f}".format(logreg.score(X_test,Y_test)))