Skip to content

Instantly share code, notes, and snippets.

What would you like to do?
Horovod PyTorch
import argparse
import os
from filelock import FileLock
import torch.multiprocessing as mp
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
from torchvision import datasets, transforms
import horovod.torch as hvd
new_mirror = ''
datasets.MNIST.resources = [
('/'.join([new_mirror, url.split('/')[-1]]), md5)
for url, md5 in datasets.MNIST.resources
# 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=42, metavar='S',
help='random seed (default: 42)')
parser.add_argument('--log-interval', type=int, default=10, metavar='N',
help='how many batches to wait before logging training status')
parser.add_argument('--fp16-allreduce', action='store_true', default=False,
help='use fp16 compression during allreduce')
parser.add_argument('--use-adasum', action='store_true', default=False,
help='use adasum algorithm to do reduction')
parser.add_argument('--gradient-predivide-factor', type=float, default=1.0,
help='apply gradient predivide factor in optimizer (default: 1.0)')
help='location of the training dataset in the local filesystem (will be downloaded if needed)')
class Net(nn.Module):
def __init__(self):
super(Net, self).__init__()
self.conv1 = nn.Conv2d(1, 10, kernel_size=5)
self.conv2 = nn.Conv2d(10, 20, kernel_size=5)
self.conv2_drop = nn.Dropout2d()
self.fc1 = nn.Linear(320, 50)
self.fc2 = nn.Linear(50, 10)
def forward(self, x):
x = F.relu(F.max_pool2d(self.conv1(x), 2))
x = F.relu(F.max_pool2d(self.conv2_drop(self.conv2(x)), 2))
x = x.view(-1, 320)
x = F.relu(self.fc1(x))
x = F.dropout(x,
x = self.fc2(x)
return F.log_softmax(x)
def train(epoch):
# Horovod: set epoch to sampler for shuffling.
for batch_idx, (data, target) in enumerate(train_loader):
if args.cuda:
data, target = data.cuda(), target.cuda()
output = model(data)
loss = F.nll_loss(output, target)
if batch_idx % args.log_interval == 0:
# Horovod: use train_sampler to determine the number of examples in
# this worker's partition.
print('Train Epoch: {} [{}/{} ({:.0f}%)]\tLoss: {:.6f}'.format(
epoch, batch_idx * len(data), len(train_sampler),
100. * batch_idx / len(train_loader), loss.item()))
def metric_average(val, name):
tensor = torch.tensor(val)
avg_tensor = hvd.allreduce(tensor, name=name)
return avg_tensor.item()
def test():
test_loss = 0.
test_accuracy = 0.
for data, target in test_loader:
if args.cuda:
data, target = data.cuda(), target.cuda()
output = model(data)
# sum up batch loss
test_loss += F.nll_loss(output, target, size_average=False).item()
# get the index of the max log-probability
pred =, keepdim=True)[1]
test_accuracy += pred.eq(
# Horovod: use test_sampler to determine the number of examples in
# this worker's partition.
test_loss /= len(test_sampler)
test_accuracy /= len(test_sampler)
# Horovod: average metric values across workers.
test_loss = metric_average(test_loss, 'avg_loss')
test_accuracy = metric_average(test_accuracy, 'avg_accuracy')
# Horovod: print output only on first rank.
if hvd.rank() == 0:
print('\nTest set: Average loss: {:.4f}, Accuracy: {:.2f}%\n'.format(
test_loss, 100. * test_accuracy))
if __name__ == '__main__':
args = parser.parse_args()
args.cuda = not args.no_cuda and torch.cuda.is_available()
# Horovod: initialize library.
if args.cuda:
# Horovod: pin GPU to local rank.
# Horovod: limit # of CPU threads to be used per worker.
kwargs = {'num_workers': 1, 'pin_memory': True} if args.cuda else {}
# When supported, use 'forkserver' to spawn dataloader workers instead of 'fork' to prevent
# issues with Infiniband implementations that are not fork-safe
if (kwargs.get('num_workers', 0) > 0 and hasattr(mp, '_supports_context') and
mp._supports_context and 'forkserver' in mp.get_all_start_methods()):
kwargs['multiprocessing_context'] = 'forkserver'
data_dir = args.data_dir or './data'
with FileLock(os.path.expanduser("~/.horovod_lock")):
train_dataset = \
datasets.MNIST(data_dir, train=True, download=True,
transforms.Normalize((0.1307,), (0.3081,))
# Horovod: use DistributedSampler to partition the training data.
train_sampler =
train_dataset, num_replicas=hvd.size(), rank=hvd.rank())
train_loader =
train_dataset, batch_size=args.batch_size, sampler=train_sampler, **kwargs)
test_dataset = \
datasets.MNIST(data_dir, train=False, transform=transforms.Compose([
transforms.Normalize((0.1307,), (0.3081,))
# Horovod: use DistributedSampler to partition the test data.
test_sampler =
test_dataset, num_replicas=hvd.size(), rank=hvd.rank())
test_loader =, batch_size=args.test_batch_size,
sampler=test_sampler, **kwargs)
model = Net()
# By default, Adasum doesn't need scaling up learning rate.
lr_scaler = hvd.size() if not args.use_adasum else 1
if args.cuda:
# Move model to GPU.
# If using GPU Adasum allreduce, scale learning rate by local_size.
if args.use_adasum and hvd.nccl_built():
lr_scaler = hvd.local_size()
# Horovod: scale learning rate by lr_scaler.
optimizer = optim.SGD(model.parameters(), * lr_scaler,
# Horovod: broadcast parameters & optimizer state.
hvd.broadcast_parameters(model.state_dict(), root_rank=0)
hvd.broadcast_optimizer_state(optimizer, root_rank=0)
# Horovod: (optional) compression algorithm.
compression = hvd.Compression.fp16 if args.fp16_allreduce else hvd.Compression.none
# Horovod: wrap optimizer with DistributedOptimizer.
optimizer = hvd.DistributedOptimizer(optimizer,
op=hvd.Adasum if args.use_adasum else hvd.Average,
for epoch in range(1, args.epochs + 1):
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment