OSS_Project2/template.py at master · fivedasol/OSS_Project2 · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
#PLEASE WRITE THE GITHUB URL BELOW!
# https://github.com/fivedasol/OSS_Project2.git

import sys
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.tree import DecisionTreeClassifier
from sklearn.svm import SVC
from sklearn.ensemble import RandomForestClassifier
from sklearn.metrics import accuracy_score, precision_score, recall_score


def load_dataset(dataset_path):
	#To-Do: Implement this function
	pop = pd.read_csv(dataset_path)
	frame = pd.DataFrame(pop)
	return frame

def dataset_stat(dataset_df):
	#To-Do: Implement this function
	n_feats = dataset_df.shape[1] - 1
	n_class0 = len(dataset_df.loc[dataset_df['target'] == 0])
	n_class1 = len(dataset_df.loc[dataset_df['target'] == 1])
	return n_feats, n_class0, n_class1


def split_dataset(dataset_df, testset_size):
	#To-Do: Implement this function
	df_x = dataset_df.drop(['target'], axis = 1)
	df_y = dataset_df['target']
	x_train, x_test, y_train, y_test = train_test_split(df_x, df_y, test_size=testset_size)
	return x_train, x_test, y_train, y_test

def decision_tree_train_test(x_train, x_test, y_train, y_test):
	#To-Do: Implement this function
	dt_cls = DecisionTreeClassifier()
	dt_cls.fit(x_train, y_train)
	acc = accuracy_score(y_test, dt_cls.predict(x_test))
	prec = precision_score(y_test, dt_cls.predict(x_test))
	recall = recall_score(y_test, dt_cls.predict(x_test))
	return acc, prec, recall

def random_forest_train_test(x_train, x_test, y_train, y_test):
	#To-Do: Implement this function
	rf_cls = RandomForestClassifier()
	rf_cls.fit(x_train, y_train)
	acc = accuracy_score(y_test, rf_cls.predict(x_test))
	prec = precision_score(y_test, rf_cls.predict(x_test))
	recall = recall_score(y_test, rf_cls.predict(x_test))
	return acc, prec, recall

def svm_train_test(x_train, x_test, y_train, y_test):
	#To-Do: Implement this function
	svm_cls = SVC()
	svm_cls.fit(x_train, y_train)
	acc = accuracy_score(y_test, svm_cls.predict(x_test))
	prec = precision_score(y_test, svm_cls.predict(x_test))
	recall = recall_score(y_test, svm_cls.predict(x_test))
	return acc, prec, recall

def print_performances(acc, prec, recall):
	#Do not modify this function!
	print ("Accuracy: ", acc)
	print ("Precision: ", prec)
	print ("Recall: ", recall)

if __name__ == '__main__':
	#Do not modify the main script!
	data_path = sys.argv[1]
	data_df = load_dataset(data_path)

	n_feats, n_class0, n_class1 = dataset_stat(data_df)
	print ("Number of features: ", n_feats)
	print ("Number of class 0 data entries: ", n_class0)
	print ("Number of class 1 data entries: ", n_class1)

	print ("\nSplitting the dataset with the test size of ", float(sys.argv[2]))
	x_train, x_test, y_train, y_test = split_dataset(data_df, float(sys.argv[2]))

	acc, prec, recall = decision_tree_train_test(x_train, x_test, y_train, y_test)
	print ("\nDecision Tree Performances")
	print_performances(acc, prec, recall)

	acc, prec, recall = random_forest_train_test(x_train, x_test, y_train, y_test)
	print ("\nRandom Forest Performances")
	print_performances(acc, prec, recall)

	acc, prec, recall = svm_train_test(x_train, x_test, y_train, y_test)
	print ("\nSVM Performances")
	print_performances(acc, prec, recall)