original code from here
reference Using_Caffe_to_Create_a_MNIST_Dataset_Recognition_Application/codes/train.py
from __future__ import print_function
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
+import os
+import moxing as mox
class Net(nn.Module):
def __init__(self):
super(Net, self).__init__()
self.conv1 = nn.Conv2d(1, 20, 5, 1)
self.conv2 = nn.Conv2d(20, 50, 5, 1)
self.fc1 = nn.Linear(4*4*50, 500)
self.fc2 = nn.Linear(500, 10)
def forward(self, x):
x = F.relu(self.conv1(x))
x = F.max_pool2d(x, 2, 2)
x = F.relu(self.conv2(x))
x = F.max_pool2d(x, 2, 2)
x = x.view(-1, 4*4*50)
x = F.relu(self.fc1(x))
x = self.fc2(x)
return F.log_softmax(x, dim=1)
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()))
def test(args, model, device, test_loader):
model.eval()
test_loss = 0
correct = 0
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()
test_loss /= 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)))
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=0.01, metavar='LR',
help='learning rate (default: 0.01)')
parser.add_argument('--momentum', type=float, default=0.5, metavar='M',
help='SGD momentum (default: 0.5)')
parser.add_argument('--no-cuda', action='store_true', default=False,
help='disables CUDA training')
parser.add_argument('--seed', type=int, default=1, metavar='S',
help='random seed (default: 1)')
parser.add_argument('--log-interval', type=int, default=10, metavar='N',
help='how many batches to wait before logging training status')
- parser.add_argument('--save-model', action='store_true', default=False,
+ parser.add_argument('--save-model', action='store_true', default=True,
help='For Saving the current Model')
- args = parser.parse_args()
+ parser.add_argument('--train_url', type=str, help='dir to save the current Model') # obs 的输出位置
+ parser.add_argument('--data_url', type=str, help='dir for dataset') # obs 的数据位置
+ parser.add_argument('--data_local_path', type=str, default="data_set", help='local dir for dataset') # local 的数据位置
+ parser.add_argument('--model_local_path', type=str, default="results", help='local dir for dataset') # local 的输出位置
+ args, _ = parser.parse_known_args() # 只使用已知的参数
use_cuda = not args.no_cuda and torch.cuda.is_available()
+ local_dataset_url = args.data_local_path
+ if not os.path.exists(local_dataset_url):
+ os.makedirs(local_dataset_url)
+ print('local_dataset_url: ' + local_dataset_url)
+
+ if mox.file.exists(args.data_url):
+ #copy data from obs to local
+ print("data obs url exists")
+ mox.file.copy_parallel(src_url=args.data_url, dst_url=local_dataset_url) # 从桶中拷贝数据到镜像中
+
+ # model save path
+ model_local_output = args.model_local_path
+ if not os.path.exists(model_local_output):
+ os.makedirs(model_local_output)
+ print("model_local_output: " + model_local_output)
+ model_obs_output = args.train_url
+
torch.manual_seed(args.seed)
device = torch.device("cuda" if use_cuda else "cpu")
kwargs = {'num_workers': 1, 'pin_memory': True} if use_cuda else {}
train_loader = torch.utils.data.DataLoader(
- datasets.MNIST('../data', train=True, download=True,
+ datasets.MNIST(args.data_local_path, train=True, download=False,
transform=transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.1307,), (0.3081,))
])),
batch_size=args.batch_size, shuffle=True, **kwargs)
test_loader = torch.utils.data.DataLoader(
- datasets.MNIST('../data', train=False, transform=transforms.Compose([
+ datasets.MNIST(args.data_local_path, train=False, transform=transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.1307,), (0.3081,))
])),
batch_size=args.test_batch_size, shuffle=True, **kwargs)
model = Net().to(device)
optimizer = optim.SGD(model.parameters(), lr=args.lr, momentum=args.momentum)
for epoch in range(1, args.epochs + 1):
train(args, model, device, train_loader, optimizer, epoch)
test(args, model, device, test_loader)
if (args.save_model):
- torch.save(model.state_dict(),"mnist_cnn.pt")
+ torch.save(model.state_dict(),os.path.join(model_local_output, "mnist_cnn.pt"))
+
+ # copy final model from local to obs
+ model_obs_output = os.path.join(model_obs_output, "final")
+ print("model_obs_output: " + model_obs_output)
+ if not mox.file.exists(model_obs_output):
+ mox.file.make_dirs(model_obs_output)
+ mox.file.copy_parallel(src_url=model_local_output, dst_url=model_obs_output) # 将训练结果拷贝回桶中
if __name__ == '__main__':
main()