-
Notifications
You must be signed in to change notification settings - Fork 0
/
resnet_trainer.py
78 lines (64 loc) · 2.41 KB
/
resnet_trainer.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
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
import torch
import torch.nn as nn
import torch.optim as optim
from torch.utils.data import DataLoader
from torchvision import transforms
from evaluate_visualization import EvaluateVisualization
# Training class
class ResNetTrainer:
def __init__(self, model, train_dataloader, val_dataloader, test_dataloader, criterion, optimizer, device='cuda'):
self.model = model
self.train_dataloader = train_dataloader
self.val_dataloader = val_dataloader
self.test_dataloader = test_dataloader
self.criterion = criterion
self.optimizer = optimizer
self.device = device
self.evaluator = EvaluateVisualization()
def train(self, num_epochs):
train_losses = []
val_losses = []
for epoch in range(num_epochs):
# Training code...
self.model.train()
for inputs, labels in self.train_dataloader:
inputs, labels = inputs.to(self.device), labels.to(self.device)
self.optimizer.zero_grad()
outputs = self.model(inputs)
loss = self.criterion(outputs, labels)
loss.backward()
self.optimizer.step()
# Validation Code..
self.model.eval()
val_loss = 0.0
val_samples = 0
y_true = []
y_pred = []
with torch.no_grad():
for inputs, labels in self.val_dataloader:
inputs, labels = inputs.to(self.device), labels.to(self.device)
outputs = self.model(inputs)
loss = self.criterion(outputs, labels)
val_loss += loss.item() * inputs.size(0)
val_samples += inputs.size(0)
_, predicted = torch.max(outputs, 1)
y_true.extend(labels.cpu().numpy())
y_pred.extend(predicted.cpu().numpy())
val_losses.append(val_loss / val_samples)
# Plotting loss curve
self.evaluator.plot_loss_curve(train_losses, val_losses)
def evaluate(self):
self.model.eval()
y_true_test = []
y_pred_test = []
with torch.no_grad():
for inputs, labels in self.test_dataloader:
inputs, labels = inputs.to(self.device), labels.to(self.device)
outputs = self.model(inputs)
_, predicted = torch.max(outputs, 1)
y_true_test.extend(labels.cpu().numpy())
y_pred_test.extend(predicted.cpu().numpy())
# Plotting confusion matrix
class_names = self.test_dataloader.dataset.classes
self.evaluator.plot_confusion_matrix(y_true_test, y_pred_test, class_names)
print('Evaluation finished.')