22년 5월 18일 정식으로 M1 pytorch 가 공개되었다.(Nightly 버전) [Introducing accelerated pytorch training on mac]

마침 M1, Intel iMac 을 둘 다 가지고 있고 시간도 잠시 생겼으니 간단하게 테스트해보자.

테스트 환경

1. Colab T4
2. M1 iMac 24”, 8 CPU, 8 GPU, 16GB RAM
   1. CPU version
   2. GPU(mps) version
3. Intel iMac 27”, 3.3 GHz 6 core Intel Core i5, 16GB RAM

설치 방법

• 설치페이지에서 Nightly - Mac - pip - python 을 선택해서 나오는 명령어로 설치할 수 있다. (이제 공식 지원 되니까!)

• 설치하고 버전을 확인해보았다.

  print(torch.__version__)
  # 1.13.0.dev20220522
  # 버그가 있어서 꼭 해당 버전 이상으로 설치하시기 바랍니다.
  

사용한 코드 - 3 layer CNN

import torch
from torchvision import datasets, transforms
from tqdm import tqdm

# Devices
# Colab - T4
device = 'cuda'
# M1 gpu
device = 'mps'
# M1 & Intel cpu
device = 'cpu'

# hyperparameters
learning_rate = 0.001
training_epochs = 10
batch_size = 100

# MNIST dataset
mnist_train = datasets.MNIST(root='MNIST_data/',
                             train=True,
                             transform=transforms.ToTensor(),
                             download=True)

mnist_test = datasets.MNIST(root='MNIST_data/',
                            train=False,
                            transform=transforms.ToTensor(),
                            download=True)

# dataset loader
data_loader = torch.utils.data.DataLoader(dataset=mnist_train,
                                          batch_size=batch_size,
                                          shuffle=True,
                                          drop_last=True)

# CNN Model (3 conv layers)
class CNN(torch.nn.Module):

    def __init__(self):
        super(CNN, self).__init__()
        self.layer1 = torch.nn.Sequential(
            torch.nn.Conv2d(1, 32, kernel_size=3, stride=1, padding=1),
            torch.nn.ReLU(),
            torch.nn.MaxPool2d(kernel_size=2, stride=2))
        self.layer2 = torch.nn.Sequential(
            torch.nn.Conv2d(32, 64, kernel_size=3, stride=1, padding=1),
            torch.nn.ReLU(),
            torch.nn.MaxPool2d(kernel_size=2, stride=2))        
        self.layer3 = torch.nn.Sequential(
            torch.nn.Conv2d(64, 128, kernel_size=3, stride=1, padding=1),
            torch.nn.ReLU(),
            torch.nn.MaxPool2d(kernel_size=2, stride=2))
        self.fc = torch.nn.Linear(3 * 3 * 128, 10, bias=True)

    def forward(self, x):
        out = self.layer1(x)
        out = self.layer2(out)
        out = self.layer3(out)
        out = out.view(out.size(0), -1)
        out = self.fc(out)
        return out

# instantiate CNN model
model = CNN().to(device)

# define cost/loss & optimizer
criterion = torch.nn.CrossEntropyLoss().to(device)
optimizer = torch.optim.Adam(model.parameters(), lr=learning_rate)

# train my model
total_batch = len(data_loader)
print('Learning started. It takes sometime.')
for epoch in range(training_epochs):
    avg_cost = 0

    for X, Y in tqdm(data_loader):
        X = X.to(device)
        Y = Y.to(device)

        optimizer.zero_grad()
        hypothesis = model(X)
        cost = criterion(hypothesis, Y)
        cost.backward()
        optimizer.step()

        avg_cost += cost / total_batch

    print('[Epoch: {:>4}] cost = {:>.9}'.format(epoch + 1, avg_cost))

print('Learning Finished!')

# Test model and check accuracy
with torch.no_grad():
    X_test = mnist_test.test_data.view(len(mnist_test), 1, 28, 28).float().to(device)
    Y_test = mnist_test.test_labels.to(device)

    prediction = model(X_test)
    correct_prediction = torch.argmax(prediction, 1) == Y_test
    accuracy = correct_prediction.float().mean()
    print('Accuracy:', accuracy.item())

결과

Colab - T4