-
Notifications
You must be signed in to change notification settings - Fork 1
/
evaluator.py
45 lines (33 loc) · 1.51 KB
/
evaluator.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
37
38
39
40
41
42
43
44
45
import torch
class Evaluation:
def __init__(self, predictions, labels):
self.predictions = predictions
self.labels = labels
self.tp = (predictions == 1) * (labels == 1)
self.tp_num = torch.nonzero(self.tp).shape[0]
self.tn = (predictions != 1) * (labels != 1)
self.tn_num = torch.nonzero(self.tn).shape[0]
self.fp = (predictions == 1) * (labels != 1)
self.fp_num = torch.nonzero(self.fp).shape[0]
self.fn = (predictions != 1) * (labels == 1)
self.fn_num = torch.nonzero(self.fn).shape[0]
self.total = len(labels)
self.precision = self.tp_num / (self.tp_num + self.fp_num) if self.tp_num + self.fp_num != 0 else 0
self.recall = self.tp_num / (self.tp_num + self.fn_num) if self.tp_num + self.fn_num != 0 else 0
def get_fp(self):
return torch.nonzero(self.fp).squeeze()
def get_tp(self):
return torch.nonzero(self.tp).squeeze()
def get_tn(self):
return torch.nonzero(self.tn).squeeze()
def get_fn(self):
return torch.nonzero(self.fn).squeeze()
def get_accuracy(self):
return (self.tp_num + self.tn_num) / self.total
def get_precision(self):
return self.precision
def get_recall(self):
return self.recall
def get_f1(self):
return 2 * self.precision * self.recall / (self.precision + self.recall) if (
self.precision + self.recall) > 0 else 0