abci_tutorial/multi_gpu/main.py at main · kenoharada/abci_tutorial · 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
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
import argparse
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
from torchvision import datasets, transforms
from torch.optim.lr_scheduler import StepLR
from models import SimpleNet
from data import set_up_data
import wandb


def train(args, model, device, train_loader, optimizer, epoch):
    model.train()
    for batch_idx, (data, target) in enumerate(train_loader):
        data, target = data.to(device), target.to(device)
        optimizer.zero_grad()
        output = model(data)
        loss = F.nll_loss(output, target)
        loss.backward()
        optimizer.step()
        if batch_idx % args.log_interval == 0:
            print('Train Epoch: {} [{}/{} ({:.0f}%)]\tLoss: {:.6f}'.format(
                epoch, batch_idx * len(data), len(train_loader.dataset),
                100. * batch_idx / len(train_loader), loss.item()))
            wandb.log({
                "Train Loss": loss.item()})
            if args.check:
                break


def test(args, model, device, test_loader, epoch):
    model.eval()
    test_loss = 0
    correct = 0

    example_images = []
    with torch.no_grad():
        for data, target in test_loader:
            data, target = data.to(device), target.to(device)
            output = model(data)
            test_loss += F.nll_loss(output, target, reduction='sum').item()  # sum up batch loss
            pred = output.argmax(dim=1, keepdim=True)  # get the index of the max log-probability
            correct += pred.eq(target.view_as(pred)).sum().item()
            example_images.append(wandb.Image(
                data[0], caption="Pred: {} Truth: {}".format(pred[0].item(), target[0])))
            if args.check:
                break

    test_loss /= len(test_loader.dataset)
    acc = 100. * correct / len(test_loader.dataset)
    print('\nTest set: Average loss: {:.4f}, Accuracy: {}/{} ({:.0f}%)\n'.format(
        test_loss, correct, len(test_loader.dataset),
        100. * correct / len(test_loader.dataset)))
    wandb.log({
        "Test Examples": example_images,
        "test_accuracy": 100. * correct / len(test_loader.dataset),
        "Test Loss": test_loss,
        "epoch": epoch})
    return acc


def main():
    # Training settings
    parser = argparse.ArgumentParser(description='PyTorch MNIST Example')
    parser.add_argument('--batch-size', type=int, default=64, metavar='N',
                        help='input batch size for training (default: 64)')
    parser.add_argument('--test-batch-size', type=int, default=1000, metavar='N',
                        help='input batch size for testing (default: 1000)')
    parser.add_argument('--epochs', type=int, default=14, metavar='N',
                        help='number of epochs to train (default: 14)')
    parser.add_argument('--lr', type=float, default=1.0, metavar='LR',
                        help='learning rate (default: 1.0)')
    parser.add_argument('--gamma', type=float, default=0.7, metavar='M',
                        help='Learning rate step gamma (default: 0.7)')
    parser.add_argument('--check', action='store_true', default=False,
                        help='check train, test, save procedure')
    parser.add_argument('--seed', type=int, default=1, metavar='S',
                        help='random seed (default: 1)')
    parser.add_argument('--log-interval', type=int, default=100, metavar='N',
                        help='how many batches to wait before logging training status')
    args = parser.parse_args()

    torch.manual_seed(args.seed)

    device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

    train_loader, test_loader = set_up_data(args)

    model = SimpleNet().to(device)
    optimizer = optim.Adadelta(model.parameters(), lr=args.lr)
    wandb.init(project="SimpleNet-mnist")
    wandb.config.update(args)
    scheduler = StepLR(optimizer, step_size=1, gamma=args.gamma)
    best_acc = 0.
    wandb.watch(model)
    for epoch in range(1, args.epochs + 1):
        train(args, model, device, train_loader, optimizer, epoch)
        acc = test(args, model, device, test_loader, epoch)
        scheduler.step()
    if best_acc < acc:
        torch.save(model.state_dict(), "mnist_cnn.pt")
        wandb.save("mnist_cnn.pt")
        best_acc = acc
        wandb.run.summary["best_acc"] = acc
    wandb.finish()


if __name__ == '__main__':
    main()