task_1
import numpy as n
from sklearn.model_selection import train_test_split
dataset=[[1,0,0,1],[1,0,0,1],[1,1,0,1],[1,1,1,0]]
W=[w0,w1,w2]=[0,0,0]
threshold=0.5
learning_rate=0.1
addition=[]
x_train_data,x_test_data=train_test_split(dataset,test_size=0.2,random_state=1)
print('x train',x_train_data)
print('x test',x_test_data)
error=[]
iterations=5
label=[]
f=[]
classes=[]
feature=[]
feature_matrix=[]
total_classes=[]
predicted_class=[]
accuracy_training=0
train_data_p=[]
weight=[]
for iter in range(iterations=5):
for sample in x_train_data:
feature=sample[0:len(x_train_data)-1]
classes = sample[-1]
o=n.dot(W,feature)
if o>=threshold:
o=1
predicted_class=o
else:
o=0
predicted_class=o
error=(learning_rate)*(classes-o)
for j in range(len(W)):
W[j]=W[j]+(feature[j]*(error))
print('W',W)
if W not in weight:
weight.append(update_w)
else:
print('UPDATED WEIGHT = ', W)
break
for i in x_test_data:
x = i[0:3]
y = i[-1]
o2=n.dot(update_w,x)
if o2 >= threshold:
o2 = 1
predict_train_data = o2
if(predict_train_data==y):
print("predicted")
else:
print("not pred")
else:
o2 = 0
predict_train_data = o2
if(predict_train_data==y):
print("predicted")
else:
print("not pred")
for k in x_train_data:
s=k[0:3]
z=k[-1]
o1 = n.dot(update_w, s)
if o1 >= threshold:
o1 = 1
predict_train_data = o1
train_data_p.append(predict_train_data)
if predict_train_data==classes:
accuracy_training = accuracy_training + 1
else:
accuracy_training = accuracy_training + 0
else:
o1 = 0
predict_train_data = o1
train_data_p.append(predict_train_data)
if predict_train_data==classes:
accuracy_training = accuracy_training + 1
else:
accuracy_training = accuracy_training + 0
if (accuracy_training/len(train_data_p))==1:
print('accurate pred')
break
else:
continue
break
