mnist_pl.py

# Based on: https://github.com/pytorch/examples/blob/master/mnist/main.py
import argparse
from pytorch_lightning import callbacks
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
import pytorch_lightning as pl


class Net(nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.conv1 = nn.Conv2d(1, 32, 3, 1)
        self.conv2 = nn.Conv2d(32, 64, 3, 1)
        self.dropout1 = nn.Dropout(0.25)
        self.dropout2 = nn.Dropout(0.5)
        self.fc1 = nn.Linear(9216, 128)
        self.fc2 = nn.Linear(128, 10)

    def forward(self, x):
        x = self.conv1(x)
        x = F.relu(x)
        x = self.conv2(x)
        x = F.relu(x)
        x = F.max_pool2d(x, 2)
        x = self.dropout1(x)
        x = torch.flatten(x, 1)
        x = self.fc1(x)
        x = F.relu(x)
        x = self.dropout2(x)
        x = self.fc2(x)
        output = F.log_softmax(x, dim=1)
        return output


class LitClassifier(pl.LightningModule):
    def __init__(self, args):
        super().__init__()
        self.args = args
        self.save_hyperparameters()
        self.model = Net()

    def forward(self, x):
        x = self.model(x)
        return x

    def training_step(self, batch, batch_idx):
        x, y = batch
        y_hat = self(x)
        loss = F.nll_loss(y_hat, y)
        return loss

    def validation_step(self, batch, batch_idx):
        x, y = batch
        logits = self(x)
        acc = self.accuracy(logits, y)
        return acc

    def test_step(self, batch, batch_idx):
        x, y = batch
        logits = self(x)
        acc = self.accuracy(logits, y)
        return acc

    def accuracy(self, logits, y):
        acc = torch.sum(torch.eq(torch.argmax(logits, -1), y).to(torch.float32)) / len(y)
        return acc

    def validation_epoch_end(self, outputs) -> None:
        self.log("val_acc", torch.stack(outputs).mean(), prog_bar=True)

    def test_epoch_end(self, outputs) -> None:
        self.log("test_acc", torch.stack(outputs).mean())

    def configure_optimizers(self):
        optimizer = optim.Adadelta(self.model.parameters(), lr=self.args.lr)
        scheduler = StepLR(optimizer, step_size=1, gamma=self.args.gamma)
        return [optimizer], [scheduler]


class ProfCallback(pl.Callback):
    def __init__(self, prof):
        self.prof = prof

    def on_train_batch_end(self, *args, **kwargs):
        self.prof.step()


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=10, metavar='N',
                        help='number of epochs to train (default: 10)')
    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('--no-cuda', action='store_true', default=False,
                        help='disables CUDA training')
    parser.add_argument('--dry-run', action='store_true', default=False,
                        help='quickly check a single pass')
    parser.add_argument('--seed', type=int, default=1, metavar='S',
                        help='random seed (default: 1)')
    parser.add_argument('--save-model', action='store_true', default=False,
                        help='For Saving the current Model')
    args = parser.parse_args()
    print(args)
    
    use_cuda = not args.no_cuda and torch.cuda.is_available()

    torch.manual_seed(args.seed)

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

    train_kwargs = {'batch_size': args.batch_size}
    test_kwargs = {'batch_size': args.test_batch_size}
    if use_cuda:
        cuda_kwargs = {'num_workers': 2,
                       'pin_memory': True,
                       'shuffle': True}
        train_kwargs.update(cuda_kwargs)

        cuda_kwargs = {'num_workers': 2,
                       'pin_memory': True,
                       'shuffle': False}
        test_kwargs.update(cuda_kwargs)

    transform=transforms.Compose([
        transforms.ToTensor(),
        transforms.Normalize((0.1307,), (0.3081,))
        ])
    dataset1 = datasets.MNIST('../data', train=True, download=True,
                       transform=transform)
    dataset2 = datasets.MNIST('../data', train=False,
                       transform=transform)
    train_loader = torch.utils.data.DataLoader(dataset1,**train_kwargs)
    test_loader = torch.utils.data.DataLoader(dataset2, **test_kwargs)

    model = LitClassifier(args)

    with torch.profiler.profile(activities=[torch.profiler.ProfilerActivity.CPU,
                                            torch.profiler.ProfilerActivity.CUDA], 
                                schedule=torch.profiler.schedule(wait=1, warmup=2, active=3, repeat=1),
                                on_trace_ready=torch.profiler.tensorboard_trace_handler('mnist-pl'),
                                profile_memory=True,
                                with_stack=False,
                                record_shapes=True) as prof:

        trainer = pl.Trainer(max_epochs=args.epochs, strategy='ddp', accelerator='gpu', gpus=4, callbacks=[ProfCallback(prof)])
        trainer.fit(model, train_loader, test_loader)
        trainer.test(dataloaders=test_loader)


if __name__ == '__main__':
    main()