diff --git a/timm/data/dataset.py b/timm/data/dataset.py index 6a71db41..93b429e0 100644 --- a/timm/data/dataset.py +++ b/timm/data/dataset.py @@ -4,6 +4,7 @@ Hacked together by / Copyright 2019, Ross Wightman """ import io import logging +from typing import Optional import torch import torch.utils.data as data @@ -132,6 +133,14 @@ class IterableImageDataset(data.IterableDataset): if hasattr(self.parser, 'set_epoch'): self.parser.set_epoch(count) + def set_loader_cfg( + self, + num_workers: Optional[int] = None, + ): + # TFDS and WDS readers need # workers for correct # samples estimate before loader processes created + if hasattr(self.parser, 'set_loader_cfg'): + self.parser.set_loader_cfg(num_workers=num_workers) + def filename(self, index, basename=False, absolute=False): assert False, 'Filename lookup by index not supported, use filenames().' diff --git a/timm/data/loader.py b/timm/data/loader.py index 35ccd503..1a4800f8 100644 --- a/timm/data/loader.py +++ b/timm/data/loader.py @@ -16,12 +16,12 @@ import torch import torch.utils.data import numpy as np -from .transforms_factory import create_transform from .constants import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD +from .dataset import IterableImageDataset from .distributed_sampler import OrderedDistributedSampler, RepeatAugSampler from .random_erasing import RandomErasing from .mixup import FastCollateMixup - +from .transforms_factory import create_transform _logger = logging.getLogger(__name__) @@ -248,6 +248,11 @@ def create_loader( separate=num_aug_splits > 0, ) + if isinstance(dataset, IterableImageDataset): + # give Iterable datasets early knowledge of num_workers so that sample estimates + # are correct before worker processes are launched + dataset.set_loader_cfg(num_workers=num_workers) + sampler = None if distributed and not isinstance(dataset, torch.utils.data.IterableDataset): if is_training: diff --git a/timm/data/parsers/parser_hfds.py b/timm/data/parsers/parser_hfds.py index bd816f8c..a558aaf3 100644 --- a/timm/data/parsers/parser_hfds.py +++ b/timm/data/parsers/parser_hfds.py @@ -49,7 +49,7 @@ class ParserHfds(Parser): self.class_to_idx = get_class_labels(self.dataset.info) self.split_info = self.dataset.info.splits[split] - self.num_examples = self.split_info.num_examples + self.num_samples = self.split_info.num_examples def __getitem__(self, index): item = self.dataset[index] diff --git a/timm/data/parsers/parser_tfds.py b/timm/data/parsers/parser_tfds.py index dd16b87a..f55f012f 100644 --- a/timm/data/parsers/parser_tfds.py +++ b/timm/data/parsers/parser_tfds.py @@ -8,6 +8,7 @@ Hacked together by / Copyright 2020 Ross Wightman """ import math import os +from typing import Optional import torch import torch.distributed as dist @@ -38,12 +39,12 @@ from .shared_count import SharedCount MAX_TP_SIZE = os.environ.get('TFDS_TP_SIZE', 8) # maximum TF threadpool size, for jpeg decodes and queuing activities -SHUFFLE_SIZE = os.environ.get('TFDS_SHUFFLE_SIZE', 8192) # examples to shuffle in DS queue -PREFETCH_SIZE = os.environ.get('TFDS_PREFETCH_SIZE', 2048) # examples to prefetch +SHUFFLE_SIZE = os.environ.get('TFDS_SHUFFLE_SIZE', 8192) # samples to shuffle in DS queue +PREFETCH_SIZE = os.environ.get('TFDS_PREFETCH_SIZE', 2048) # samples to prefetch -def even_split_indices(split, n, num_examples): - partitions = [round(i * num_examples / n) for i in range(n + 1)] +def even_split_indices(split, n, num_samples): + partitions = [round(i * num_samples / n) for i in range(n + 1)] return [f"{split}[{partitions[i]}:{partitions[i + 1]}]" for i in range(n)] @@ -59,20 +60,20 @@ class ParserTfds(Parser): """ Wrap Tensorflow Datasets for use in PyTorch There several things to be aware of: - * To prevent excessive examples being dropped per epoch w/ distributed training or multiplicity of + * To prevent excessive samples being dropped per epoch w/ distributed training or multiplicity of dataloader workers, the train iterator wraps to avoid returning partial batches that trigger drop_last https://github.com/pytorch/pytorch/issues/33413 * With PyTorch IterableDatasets, each worker in each replica operates in isolation, the final batch from each worker could be a different size. For training this is worked around by option above, for - validation extra examples are inserted iff distributed mode is enabled so that the batches being reduced + validation extra samples are inserted iff distributed mode is enabled so that the batches being reduced across replicas are of same size. This will slightly alter the results, distributed validation will not be 100% correct. This is similar to common handling in DistributedSampler for normal Datasets but a bit worse - since there are up to N * J extra examples with IterableDatasets. + since there are up to N * J extra samples with IterableDatasets. * The sharding (splitting of dataset into TFRecord) files imposes limitations on the number of replicas and dataloader workers you can use. For really small datasets that only contain a few shards you may have to train non-distributed w/ 1-2 dataloader workers. This is likely not a huge concern as the benefit of distributed training or fast dataloading should be much less for small datasets. - * This wrapper is currently configured to return individual, decompressed image examples from the TFDS + * This wrapper is currently configured to return individual, decompressed image samples from the TFDS dataset. The augmentation (transforms) and batching is still done in PyTorch. It would be possible to specify TF augmentation fn and return augmented batches w/ some modifications to other downstream components. @@ -104,7 +105,7 @@ class ParserTfds(Parser): name: tfds dataset name (eg `imagenet2012`) split: tfds dataset split (can use all TFDS split strings eg `train[:10%]`) is_training: training mode, shuffle enabled, dataset len rounded by batch_size - batch_size: batch_size to use to unsure total examples % batch_size == 0 in training across all dis nodes + batch_size: batch_size to use to unsure total samples % batch_size == 0 in training across all dis nodes download: download and build TFDS dataset if set, otherwise must use tfds CLI repeats: iterate through (repeat) the dataset this many times per iteration (once if 0 or 1) seed: common seed for shard shuffle across all distributed/worker instances @@ -143,7 +144,7 @@ class ParserTfds(Parser): self.builder.download_and_prepare() self.class_to_idx = get_class_labels(self.builder.info) if self.target_name == 'label' else {} self.split_info = self.builder.info.splits[split] - self.num_examples = self.split_info.num_examples + self.num_samples = self.split_info.num_examples # Distributed world state self.dist_rank = 0 @@ -154,6 +155,7 @@ class ParserTfds(Parser): # Attributes that are updated in _lazy_init, including the tf.data pipeline itself self.global_num_workers = 1 + self.num_workers = 1 self.worker_info = None self.worker_seed = 0 # seed unique to each work instance self.subsplit = None # set when data is distributed across workers using sub-splits @@ -167,6 +169,16 @@ class ParserTfds(Parser): def set_epoch(self, count): self.epoch_count.value = count + def set_loader_cfg( + self, + num_workers: Optional[int] = None, + ): + if self.ds is not None: + return + if num_workers is not None: + self.num_workers = num_workers + self.global_num_workers = self.dist_num_replicas * self.num_workers + def _lazy_init(self): """ Lazily initialize the dataset. @@ -186,9 +198,9 @@ class ParserTfds(Parser): if worker_info is not None: self.worker_info = worker_info self.worker_seed = worker_info.seed - num_workers = worker_info.num_workers - self.global_num_workers = self.dist_num_replicas * num_workers - global_worker_id = self.dist_rank * num_workers + worker_info.id + self.num_workers = worker_info.num_workers + self.global_num_workers = self.dist_num_replicas * self.num_workers + global_worker_id = self.dist_rank * self.num_workers + worker_info.id """ Data sharding InputContext will assign subset of underlying TFRecord files to each 'pipeline' if used. @@ -198,17 +210,17 @@ class ParserTfds(Parser): I am currently using a mix of InputContext shard assignment and fine-grained sub-splits for distributing the data across workers. For training InputContext is used to assign shards to nodes unless num_shards in dataset < total number of workers. Otherwise sub-split API is used for datasets without enough shards or - for validation where we can't drop examples and need to avoid minimize uneven splits to avoid padding. + for validation where we can't drop samples and need to avoid minimize uneven splits to avoid padding. """ should_subsplit = self.global_num_workers > 1 and ( self.split_info.num_shards < self.global_num_workers or not self.is_training) if should_subsplit: - # split the dataset w/o using sharding for more even examples / worker, can result in less optimal + # split the dataset w/o using sharding for more even samples / worker, can result in less optimal # read patterns for distributed training (overlap across shards) so better to use InputContext there if has_buggy_even_splits: # my even_split workaround doesn't work on subsplits, upgrade tfds! if not isinstance(self.split_info, tfds.core.splits.SubSplitInfo): - subsplits = even_split_indices(self.split, self.global_num_workers, self.num_examples) + subsplits = even_split_indices(self.split, self.global_num_workers, self.num_samples) self.subsplit = subsplits[global_worker_id] else: subsplits = tfds.even_splits(self.split, self.global_num_workers) @@ -235,7 +247,7 @@ class ParserTfds(Parser): # avoid overloading threading w/ combo of TF ds threads + PyTorch workers options = tf.data.Options() thread_member = 'threading' if hasattr(options, 'threading') else 'experimental_threading' - getattr(options, thread_member).private_threadpool_size = max(1, self.max_threadpool_size // num_workers) + getattr(options, thread_member).private_threadpool_size = max(1, self.max_threadpool_size // self.num_workers) getattr(options, thread_member).max_intra_op_parallelism = 1 ds = ds.with_options(options) if self.is_training or self.repeats > 1: @@ -243,60 +255,65 @@ class ParserTfds(Parser): # see warnings at https://pytorch.org/docs/stable/data.html#multi-process-data-loading ds = ds.repeat() # allow wrap around and break iteration manually if self.is_training: - ds = ds.shuffle(min(self.num_examples, self.shuffle_size) // self.global_num_workers, seed=self.worker_seed) - ds = ds.prefetch(min(self.num_examples // self.global_num_workers, self.prefetch_size)) + ds = ds.shuffle(min(self.num_samples, self.shuffle_size) // self.global_num_workers, seed=self.worker_seed) + ds = ds.prefetch(min(self.num_samples // self.global_num_workers, self.prefetch_size)) self.ds = tfds.as_numpy(ds) self.init_count += 1 + def _num_samples_per_worker(self): + num_worker_samples = \ + max(1, self.repeats) * self.num_samples / max(self.global_num_workers, self.dist_num_replicas) + if self.is_training or self.dist_num_replicas > 1: + num_worker_samples = math.ceil(num_worker_samples) + if self.is_training and self.batch_size is not None: + num_worker_samples = math.ceil(num_worker_samples / self.batch_size) * self.batch_size + return int(num_worker_samples) + def __iter__(self): if self.ds is None or self.reinit_each_iter: self._lazy_init() # Compute a rounded up sample count that is used to: # 1. make batches even cross workers & replicas in distributed validation. - # This adds extra examples and will slightly alter validation results. + # This adds extra samples and will slightly alter validation results. # 2. determine loop ending condition in training w/ repeat enabled so that only full batch_size # batches are produced (underlying tfds iter wraps around) - target_example_count = math.ceil(max(1, self.repeats) * self.num_examples / self.global_num_workers) - if self.is_training: - # round up to nearest batch_size per worker-replica - target_example_count = math.ceil(target_example_count / self.batch_size) * self.batch_size + target_sample_count = self._num_samples_per_worker() # Iterate until exhausted or sample count hits target when training (ds.repeat enabled) - example_count = 0 - for example in self.ds: - input_data = example[self.input_name] + sample_count = 0 + for sample in self.ds: + input_data = sample[self.input_name] if self.input_img_mode: input_data = Image.fromarray(input_data, mode=self.input_img_mode) - target_data = example[self.target_name] + target_data = sample[self.target_name] if self.target_img_mode: target_data = Image.fromarray(target_data, mode=self.target_img_mode) yield input_data, target_data - example_count += 1 - if self.is_training and example_count >= target_example_count: + sample_count += 1 + if self.is_training and sample_count >= target_sample_count: # Need to break out of loop when repeat() is enabled for training w/ oversampling - # this results in extra examples per epoch but seems more desirable than dropping + # this results in extra samples per epoch but seems more desirable than dropping # up to N*J batches per epoch (where N = num distributed processes, and J = num worker processes) break - # Pad across distributed nodes (make counts equal by adding examples) + # Pad across distributed nodes (make counts equal by adding samples) if not self.is_training and self.dist_num_replicas > 1 and self.subsplit is not None and \ - 0 < example_count < target_example_count: + 0 < sample_count < target_sample_count: # Validation batch padding only done for distributed training where results are reduced across nodes. # For single process case, it won't matter if workers return different batch sizes. # If using input_context or % based splits, sample count can vary significantly across workers and this # approach should not be used (hence disabled if self.subsplit isn't set). - while example_count < target_example_count: + while sample_count < target_sample_count: yield input_data, target_data # yield prev sample again - example_count += 1 + sample_count += 1 def __len__(self): - # this is just an estimate and does not factor in extra examples added to pad batches based on - # complete worker & replica info (not available until init in dataloader). - return math.ceil(max(1, self.repeats) * self.num_examples / self.dist_num_replicas) + num_samples = self._num_samples_per_worker() * self.num_workers + return num_samples def _filename(self, index, basename=False, absolute=False): - assert False, "Not supported" # no random access to examples + assert False, "Not supported" # no random access to samples def filenames(self, basename=False, absolute=False): """ Return all filenames in dataset, overrides base""" @@ -304,7 +321,7 @@ class ParserTfds(Parser): self._lazy_init() names = [] for sample in self.ds: - if len(names) > self.num_examples: + if len(names) > self.num_samples: break # safety for ds.repeat() case if 'file_name' in sample: name = sample['file_name'] diff --git a/timm/data/parsers/parser_wds.py b/timm/data/parsers/parser_wds.py index 8e470a38..21bfc3e2 100644 --- a/timm/data/parsers/parser_wds.py +++ b/timm/data/parsers/parser_wds.py @@ -12,7 +12,7 @@ import sys from dataclasses import dataclass from functools import partial from itertools import islice -from typing import Dict, Tuple +from typing import Any, Callable, Dict, List, Optional, Tuple import torch import torch.distributed as dist @@ -345,6 +345,16 @@ class ParserWds(Parser): def set_epoch(self, count): self.epoch_count.value = count + def set_loader_cfg( + self, + num_workers: Optional[int] = None, + ): + if self.ds is not None: + return + if num_workers is not None: + self.num_workers = num_workers + self.global_num_workers = self.dist_num_replicas * self.num_workers + def _lazy_init(self): """ Lazily initialize worker (in worker processes) """ @@ -396,25 +406,27 @@ class ParserWds(Parser): for s in src: yield s + def _num_samples_per_worker(self): + num_worker_samples = self.num_samples / max(self.global_num_workers, self.dist_num_replicas) + if self.is_training or self.dist_num_replicas > 1: + num_worker_samples = math.ceil(num_worker_samples) + if self.is_training and self.batch_size is not None: + num_worker_samples = math.ceil(num_worker_samples / self.batch_size) * self.batch_size + return int(num_worker_samples) + def __iter__(self): if self.ds is None: self._lazy_init() - if self.is_training: - num_worker_samples = math.floor(self.num_samples / self.global_num_workers) - if self.batch_size is not None: - num_worker_samples = (num_worker_samples // self.batch_size) * self.batch_size + num_worker_samples = self._num_samples_per_worker() + if self.is_training or self.dist_num_replicas > 1: + # NOTE: doing distributed validation w/ WDS is messy, hard to meet constraints that + # same # of batches needed across all replicas w/ seeing each sample once. + # with_epoch() is simple but could miss a shard's worth of samples in some workers, + # and duplicate in others. Best to keep num DL workers low and a divisor of #val shards. ds = self.ds.with_epoch(num_worker_samples) else: - if self.dist_num_replicas > 1: - # doing distributed validation w/ WDS is messy, hard to meet constraints that - # same # of batches needed across all replicas w/ seeing each sample once. - # with_epoch() is simple but could miss a shard's worth of samples in some workers, - # and duplicate in others. Best to keep num DL workers low and a divisor of #val shards. - num_worker_samples = math.ceil(self.num_samples / self.global_num_workers) - ds = self.ds.with_epoch(num_worker_samples) - else: - ds = self.ds + ds = self.ds i = 0 _logger.info(f'start {i}, {self.worker_id}') # FIXME temporary debug @@ -424,7 +436,8 @@ class ParserWds(Parser): _logger.info(f'end {i}, {self.worker_id}') # FIXME temporary debug def __len__(self): - return math.ceil(max(1, self.repeats) * self.num_samples / self.dist_num_replicas) + num_samples = self._num_samples_per_worker() * self.num_workers + return num_samples def _filename(self, index, basename=False, absolute=False): assert False, "Not supported" # no random access to examples