File size: 2,185 Bytes
b4d7ac8 |
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 |
import torch
import math
import numpy as np
class SequentialDistributedSampler(torch.utils.data.sampler.Sampler):
"""
Distributed Sampler that subsamples indicies sequentially,
making it easier to collate all results at the end.
Even though we only use this sampler for eval and predict (no training),
which means that the model params won't have to be synced (i.e. will not hang
for synchronization even if varied number of forward passes), we still add extra
samples to the sampler to make it evenly divisible (like in `DistributedSampler`)
to make it easy to `gather` or `reduce` resulting tensors at the end of the loop.
"""
def __init__(self, dataset, batch_size, rank=None, num_replicas=None):
if num_replicas is None:
if not torch.distributed.is_available():
raise RuntimeError("Requires distributed package to be available")
num_replicas = torch.distributed.get_world_size()
if rank is None:
if not torch.distributed.is_available():
raise RuntimeError("Requires distributed package to be available")
rank = torch.distributed.get_rank()
self.dataset = dataset
self.num_replicas = num_replicas
self.rank = rank
self.batch_size = batch_size
self.num_samples = int(math.ceil(len(self.dataset) * 1.0 / self.batch_size / self.num_replicas)) * self.batch_size
self.total_size = self.num_samples * self.num_replicas
def __iter__(self):
indices = list(range(len(self.dataset)))
# add extra samples to make it evenly divisible
indices += [indices[-1]] * (self.total_size - len(indices))
# subsample
indices = indices[self.rank * self.num_samples : (self.rank + 1) * self.num_samples]
return iter(indices)
def __len__(self):
return self.num_samples
def distributed_concat(tensor, num_total_examples):
output_tensors = [tensor.clone() for _ in range(torch.distributed.get_world_size())]
torch.distributed.all_gather(output_tensors, tensor)
concat = torch.cat(output_tensors, dim=0)
return concat[:num_total_examples] |