-
Notifications
You must be signed in to change notification settings - Fork 0
/
multiclass_logistic_regression.py
36 lines (32 loc) · 1.23 KB
/
multiclass_logistic_regression.py
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
from logistic_regression import LogisticRegression
import numpy as np
class MulticlassLogisticRegression:
def __init__(self, number_of_classes, number_of_iterations, learning_rate):
"""
:param number_of_classes: number of classes to predict, must be greater than 2.
:param number_of_iterations: number of iterations of each classifier.
:param learning_rate: learning rate.
"""
assert (number_of_classes > 2)
self.nc = number_of_classes
self.lrs = []
for i in range(0, self.nc):
self.lrs.append(LogisticRegression(number_of_iterations, learning_rate))
def fit(self, X, y):
"""
:param X: input.
:param y: desired output.
"""
for i in range(0, len(self.lrs)):
self.lrs[i].fit(X, (y == i))
def predict(self, X):
"""
:param X: input.
:return: class index with maximum probability.
"""
y_prob = []
for i in range(0, len(self.lrs)):
y_prob.append(self.lrs[i].predict_prob(X))
winner_class = np.argmax(y_prob)
print("Predicted class is " + str(winner_class) + ", with probability " + str(y_prob[winner_class]))
return winner_class