ML for Beginners

ML_for_Beginners/LinearRegression/PartC.py

@@ -0,0 +1,139 @@
+import pandas as pd
+import numpy as np
+import matplotlib.pyplot as plt
+import warnings
+import math
+import copy
+from sklearn.model_selection import train_test_split, cross_val_score
+from sklearn.tree import DecisionTreeClassifier
+from sklearn.metrics import accuracy_score, confusion_matrix, roc_auc_score
+from sklearn.feature_selection import SelectFromModel
+from sklearn.cluster import KMeans
+# ======================================================================
+def gradientDescent(X, Y, learning_rate = 0.5, max_iter=1000,normal=0, lasso = 0, ridge = 0):
+	RMS = 0
+	thetas = np.zeros([1, X.shape[1]])
+	iterations=0
+	thetas_temp = np.zeros([1, X.shape[1]])
+	while iterations<(max_iter):
+		for j in range(X.shape[1]):
+			thetas_temp[0][j] = thetas[0][j]
+			if (j == 0):
+				pass
+			else:
+				if (thetas_temp[0][j] > 0):
+					check1=(1*lasso + thetas[0][j]*2*ridge)
+					thetas_temp[0][j] -= check1/(2*X.shape[0])
+				else:
+					check2=((-1)*lasso + 2*ridge*thetas[0][j])
+					thetas_temp[0][j] -= check2/(2*X.shape[0])
+		for i in range(X.shape[0]):
+			for j in range(X.shape[1]):
+				thetas_temp[0][j] -= ((learning_rate*((np.dot(thetas[0], X[i])) - Y[i]))/(X.shape[0]))*X[i][j]
+		for j in range(X.shape[1]):
+			thetas[0][j] = thetas_temp[0][j]
+		i=0
+		while (i<X.shape[0]):
+			temp1=np.dot(thetas[0], X[i])
+			temp2=temp1- Y[i]
+			RMS += math.pow(temp2,2)
+			i+=1
+		for j in range(1, X.shape[1]):
+			temp=(thetas[0][j])**2
+			check=(ridge*(temp) + lasso*(abs(thetas[0][j])))
+			RMS += check/2*X.shape[0]
+		RMS /= (2*X.shape[0])
+		iterations+=1
+	if normal==1:
+		return RMS, thetas
+	if lasso!=0:
+		return RMS,thetas
+	if ridge!=0:
+		return RMS,thetas
+
+df=pd.read_csv('data.csv')
+X = df['Brain_Weight'].values
+Y = df['Body_Weight'].values
+x1=np.c_[np.ones(X.shape[0]),X]
+# print(x1.shape)
+mean = [0]*x1.shape[1]
+maximum = [-10000000]*x1.shape[1]
+minimum = [10000000]*x1.shape[1]
+x1norm = x1.copy()
+i=0
+while i <(x1.shape[0]):
+	# only the column with actual values
+	j=1
+	mean[j] += x1norm[i][j]/x1.shape[0]
+	check=x1norm[i][j]
+	if (x1norm[i][j] > maximum[j]):
+		maximum[j] = check
+	if (x1norm[i][j] < minimum[j]):
+		minimum[j] =check
+	i+=1
+
+k=0
+while k<x1.shape[0]:
+	j=1
+	x1norm[k][j] = x1norm[k][j] - mean[j]
+	check1=maximum[j] - minimum[j]
+	x1norm[k][j] = x1norm[k][j]/(check1)
+	k+=1
+# print(x1norm)
+def graph(choice):
+	if choice==1:
+		rms, theta= gradientDescent(x1norm, Y,normal=1)
+		check5=(maximum[1] - minimum[1])
+		theta2 = theta[0][1]/check5
+		theta1 = theta[0][0] - theta2*mean[1]
+		plt.scatter(X, Y)
+		axes = plt.gca()
+		x_vals = np.array(axes.get_xlim())
+		y_vals = theta1 + theta2 * x_vals
+		plt.plot(x_vals, y_vals)
+		plt.title('Scatter plot for Normal Gradient Descent')
+		plt.xlabel('Brain Weight')
+		plt.ylabel('Body Weight')
+		plt.show()
+	elif choice==2:
+		rms, theta= gradientDescent(x1norm, Y,ridge=0.1)
+		check5=(maximum[1] - minimum[1])
+		theta2 = theta[0][1]/check5
+		theta1 = theta[0][0] - theta2*mean[1]
+		plt.scatter(X, Y)
+		axes = plt.gca()
+		x_vals = np.array(axes.get_xlim())
+		y_vals = theta1 + theta2 * x_vals
+		plt.plot(x_vals, y_vals)
+		plt.title('Scatter plot for Ridge Gradient Descent')
+		plt.xlabel('Brain Weight')
+		plt.ylabel('Body Weight')
+		plt.show()
+	else:
+		rms, theta= gradientDescent(x1norm, Y,lasso=0.7)
+		check5=(maximum[1] - minimum[1])
+		theta2 = theta[0][1]/check5
+		theta1 = theta[0][0] - theta2*mean[1]
+		plt.scatter(X, Y)
+		axes = plt.gca()
+		x_vals = np.array(axes.get_xlim())
+		y_vals = theta1 + theta2 * x_vals
+		plt.plot(x_vals, y_vals)
+		plt.title('Scatter plot for Lasso Gradient Descent')
+		plt.xlabel('Brain Weight')
+		plt.ylabel('Body Weight')
+		plt.show()
+print('Enter choice for Gradient Descent')
+print('1.Normal 2.RidgeRegularized 3.LassoRegularized' )
+input1=int(input())
+if input1==1:
+	graph(1)
+elif input1==2:
+	graph(2)
+else:
+	graph(3)
+
+# print(x1)
+# ======================================================================
+
+# ==========================================================

ML_for_Beginners/LinearRegression/README

@@ -0,0 +1 @@
+This repository has code that implements LinearRegression with regularization from scratch using python.

ML_for_Beginners/LinearRegression/data.csv

@@ -0,0 +1,168 @@
+Brain_Weight,Body_Weight
+1.0,8.429337501253315
+1.6,10.51622485105815
+2.2,12.339744042559786
+2.8000000000000003,19.217968745418617
+3.4000000000000004,19.749753099850192
+4.0,20.870851165656457
+4.6000000000000005,23.821834197373732
+5.200000000000001,28.25714507500938
+5.800000000000001,32.02395312331614
+6.4,23.72207465613001
+7.000000000000001,40.88706492165431
+7.600000000000001,32.173100245002075
+8.200000000000001,38.924977394858914
+8.8,44.784152179277044
+9.400000000000002,47.88179649821765
+10.000000000000002,42.301880217989144
+10.600000000000001,51.392283372814916
+11.200000000000001,50.02608171785642
+11.8,69.1264734682581
+12.400000000000002,43.2717139737902
+13.000000000000002,49.978282129342865
+13.600000000000001,57.36919808211232
+14.200000000000003,76.19985111659967
+14.800000000000002,87.17420938790532
+15.400000000000002,88.64977384718975
+16.0,65.42422805419027
+16.6,68.0350415440881
+17.200000000000003,98.84469923444162
+17.800000000000004,79.61612659014075
+18.400000000000002,89.90804708497112
+19.000000000000004,91.05868486914645
+19.6,91.81582543559142
+20.200000000000003,66.42501458547926
+20.800000000000004,114.1780060000467
+21.400000000000002,114.72832232041631
+22.000000000000004,92.7587433024462
+22.6,128.56238028882933
+23.200000000000003,118.79558215844521
+23.800000000000004,101.26781637686868
+24.400000000000002,110.32654535502952
+25.000000000000004,119.80000000000003
+25.600000000000005,122.55994092534604
+26.200000000000003,127.2942563207219
+26.800000000000004,128.7838874513495
+27.400000000000006,129.22661335317832
+28.000000000000004,137.23532884340943
+28.600000000000005,133.8677251419253
+29.200000000000003,132.9995176499797
+29.800000000000004,141.76247369065763
+30.400000000000006,129.96612762166652
+31.000000000000004,149.51684013184948
+31.600000000000005,157.00916281446777
+32.2,165.89173687823117
+32.800000000000004,129.6150471806473
+33.400000000000006,151.4449475556998
+34.00000000000001,162.22021707570545
+34.60000000000001,163.5908331258433
+35.2,177.85977229846404
+35.800000000000004,174.7399978102262
+36.400000000000006,173.35180569538315
+37.00000000000001,196.2845719949171
+37.60000000000001,186.0005921176151
+38.2,188.39190636090288
+38.800000000000004,197.46035734088946
+39.400000000000006,182.00480909120648
+40.00000000000001,173.6279819941465
+40.60000000000001,183.2151219808904
+41.2,175.69591117554117
+41.800000000000004,176.39770283514744
+42.400000000000006,229.6551295612347
+43.00000000000001,191.91217985668558
+43.60000000000001,230.92384733276657
+44.2,196.39550054451226
+44.800000000000004,234.59684873820942
+45.400000000000006,189.02657498675777
+46.00000000000001,218.14685141822648
+46.60000000000001,200.99133805553217
+47.20000000000001,229.8417909155177
+47.800000000000004,217.0272036042009
+48.400000000000006,231.1783698952044
+49.00000000000001,214.18980914831027
+49.60000000000001,221.60856531740188
+50.20000000000001,238.0073251669907
+50.800000000000004,240.50500555217104
+51.400000000000006,241.6163187899625
+52.00000000000001,247.4681564637112
+52.60000000000001,241.65656196087923
+53.20000000000001,253.8406575058408
+53.80000000000001,255.72640045216093
+54.400000000000006,255.9315732527974
+55.00000000000001,260.76283679315435
+55.60000000000001,263.01110536207887
+56.20000000000001,259.2065205785753
+56.80000000000001,277.9891119547556
+57.400000000000006,274.2183994005689
+58.00000000000001,281.4482136720826
+58.60000000000001,287.21717545343773
+59.20000000000001,291.86961132160087
+59.80000000000001,293.1737490742374
+60.07700002, 100.927658196754
+60.400000000000006,299.93356126189167
+61.00000000000001,291.45360012239126
+61.60000000000001,304.4382074672925
+62.20000000000001,276.3012397677959
+62.80000000000001,295.7121195288604
+63.400000000000006,305.07609455616733
+64.0,307.58703459013816
+64.60000000000001,303.1847249171093
+65.2,309.4881584965749
+65.80000000000001,297.2321629554691
+66.4,343.23277879088675
+67.00000000000001,280.47649257292863
+67.60000000000001,293.4510858685041
+68.20000000000002,321.40436855280603
+68.80000000000001,338.06711549660463
+69.4,328.1029041056382
+70.00000000000001,334.3607835363164
+70.60000000000001,349.59637439492934
+71.20000000000002,322.8907098649025
+71.80000000000001,331.6019310700573
+72.4,318.51783041479024
+73.00000000000001,318.79980576218935
+73.60000000000001,290.5488240719017
+74.20000000000002,346.2076029362015
+74.80000000000001,326.81887450112805
+75.4,356.78964794087
+76.00000000000001,358.9929135926017
+76.60000000000001,364.1137600258971
+77.20000000000002,366.0241755816051
+77.80000000000001,364.55467319794496
+78.4,365.1957441320635
+79.00000000000001,374.31985150679174
+79.60000000000001,371.38340130902884
+80.20000000000002,384.8444030232315
+80.80000000000001,381.1673354686829
+81.4,376.37417419155634
+82.00000000000001,377.1573177637777
+82.60000000000001,390.05796194250985
+83.20000000000002,410.6884609284761
+83.80000000000001,400.8694496075399
+84.4,418.0638008455297
+85.00000000000001,409.2500513592809
+85.60000000000001,407.67566308551125
+86.20000000000002,418.72819245784314
+86.80000000000001,413.39522594752384
+87.4,428.1227328827845
+88.00000000000001,432.9131332633474
+88.60000000000001,423.87711853910344
+89.20000000000002,421.38515644224384
+89.80000000000001,412.36013650301686
+90.40000000000002,435.08962625053834
+91.00000000000001,437.44750848518805
+91.60000000000001,421.15202039088297
+92.20000000000002,438.82590662840266
+92.80000000000001,449.9052452705142
+93.40000000000002,423.9039865498326
+94.00000000000001,421.3626895013748
+94.60000000000001,480.48612141637915
+95.20000000000002,443.1061433376697
+95.80000000000001,436.81613839388746
+96.40000000000002,458.01504447328927
+97.00000000000001,470.6110468364924
+97.60000000000001,430.9284381391649
+98.20000000000002,470.86835262163953
+98.80000000000001,484.8748874740794
+99.40000000000002,475.10319451317406
+100.008675894355, 120.5368757