Warn user when IterableDataset has __len__ defined (#2437)

* add warning when getting checking len

* added test

* changelog

* pep

* do not show warning below 1.4

* try version parse

* comments

* xfail

* Update requirements/base.txt

Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>

* Update pytorch_lightning/trainer/data_loading.py

Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>

* version

Co-authored-by: William Falcon <waf2107@columbia.edu>
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: Jirka <jirka@pytorchlightning.ai>

CHANGELOG.md

@@ -10,6 +10,8 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 
 - Added reduce ddp results on eval ([#2434](https://github.com/PyTorchLightning/pytorch-lightning/pull/2434))
 
+- Added a warning when an `IterableDataset` has `__len__` defined ([#2437](https://github.com/PyTorchLightning/pytorch-lightning/pull/2437))
+
 ### Changed
 
 

pytorch_lightning/trainer/data_loading.py

@@ -1,8 +1,10 @@
+import multiprocessing
 import platform
 from abc import ABC, abstractmethod
+from distutils.version import LooseVersion
 from typing import Union, List, Tuple, Callable, Optional
-import multiprocessing
 
+import torch
 import torch.distributed as torch_distrib
 from torch.utils.data import DataLoader, RandomSampler, SequentialSampler
 from torch.utils.data.distributed import DistributedSampler
@@ -41,19 +43,33 @@ else:
     HOROVOD_AVAILABLE = True
 
 
+def _has_iterable_dataset(dataloader: DataLoader):
+    return ITERABLE_DATASET_EXISTS and hasattr(dataloader, 'dataset') \
+        and isinstance(dataloader.dataset, IterableDataset)
+
+
 def _has_len(dataloader: DataLoader) -> bool:
     """ Checks if a given Dataloader has __len__ method implemented i.e. if
-    it is a finite dataloader or infinite dataloader """
+    it is a finite dataloader or infinite dataloader. """
+
     try:
         # try getting the length
         if len(dataloader) == 0:
             raise ValueError('`Dataloader` returned 0 length.'
                              ' Please make sure that your Dataloader at least returns 1 batch')
-        return True
+        has_len = True
     except TypeError:
-        return False
+        has_len = False
     except NotImplementedError:  # e.g. raised by torchtext if a batch_size_fn is used
-        return False
+        has_len = False
+
+    if has_len and _has_iterable_dataset(dataloader) and LooseVersion(torch.__version__) >= LooseVersion("1.4.0"):
+        rank_zero_warn(
+            'Your `IterableDataset` has `__len__` defined.'
+            ' In combination with multi-processing data loading (e.g. batch size > 1),'
+            ' this can lead to unintended side effects since the samples will be duplicated.'
+        )
+    return has_len
 
 
 class TrainerDataLoadingMixin(ABC):
@@ -128,12 +144,9 @@ class TrainerDataLoadingMixin(ABC):
     def auto_add_sampler(self, dataloader: DataLoader, train: bool) -> DataLoader:
 
         # don't do anything if it's not a dataloader
-        # don't manipulate iterable datasets
         is_dataloader = isinstance(dataloader, DataLoader)
-
-        is_iterable_ds = False
-        if ITERABLE_DATASET_EXISTS and hasattr(dataloader, 'dataset'):
-            is_iterable_ds = isinstance(dataloader.dataset, IterableDataset)
+        # don't manipulate iterable datasets
+        is_iterable_ds = _has_iterable_dataset(dataloader)
 
         if not is_dataloader or is_iterable_ds:
             return dataloader
@@ -285,11 +298,7 @@ class TrainerDataLoadingMixin(ABC):
         # datasets could be none, 1 or 2+
         if len(dataloaders) != 0:
             for i, dataloader in enumerate(dataloaders):
-                try:
-                    num_batches = len(dataloader)
-                except (TypeError, NotImplementedError):
-                    num_batches = float('inf')
-
+                num_batches = len(dataloader) if _has_len(dataloader) else float('inf')
                 self._worker_check(dataloader, f'{mode} dataloader {i}')
 
                 # percent or num_steps

requirements/base.txt

@@ -6,4 +6,4 @@ tensorboard>=1.14
 future>=0.17.1  # required for builtins in setup.py
 # pyyaml>=3.13
 PyYAML>=5.1  # OmegaConf requirement
-tqdm>=4.41.0
+tqdm>=4.41.0

tests/trainer/test_dataloaders.py

@@ -2,11 +2,13 @@ import platform
 
 import pytest
 import torch
+from packaging.version import parse
 from torch.utils.data.dataloader import DataLoader
-from torch.utils.data.dataset import Subset
+from torch.utils.data.dataset import Subset, IterableDataset
 
 import tests.base.develop_pipelines as tpipes
 from pytorch_lightning import Trainer
+from pytorch_lightning.trainer.data_loading import _has_len, _has_iterable_dataset
 from pytorch_lightning.utilities.exceptions import MisconfigurationException
 from tests.base import EvalModelTemplate
 
@@ -487,6 +489,36 @@ def test_warning_with_few_workers(tmpdir, ckpt_path):
         trainer.test(**test_options)
 
 
+@pytest.mark.xfail(
+    parse(torch.__version__) < parse("1.4.0"),
+    reason="IterableDataset with __len__ before 1.4 raises",
+)
+def test_warning_with_iterable_dataset_and_len(tmpdir):
+    """ Tests that a warning messages is shown when an IterableDataset defines `__len__`. """
+    model = EvalModelTemplate()
+    original_dataset = model.train_dataloader().dataset
+
+    class IterableWithLen(IterableDataset):
+
+        def __iter__(self):
+            return iter(original_dataset)
+
+        def __len__(self):
+            return len(original_dataset)
+
+    dataloader = DataLoader(IterableWithLen(), batch_size=16)
+    assert _has_len(dataloader)
+    assert _has_iterable_dataset(dataloader)
+    trainer = Trainer(
+        default_root_dir=tmpdir,
+        max_steps=3,
+    )
+    with pytest.warns(UserWarning, match='Your `IterableDataset` has `__len__` defined.'):
+        trainer.fit(model, train_dataloader=dataloader, val_dataloaders=[dataloader])
+    with pytest.warns(UserWarning, match='Your `IterableDataset` has `__len__` defined.'):
+        trainer.test(model, test_dataloaders=[dataloader])
+
+
 @pytest.mark.skipif(torch.cuda.device_count() < 2, reason='Test requires multiple GPUs')
 def test_dataloader_reinit_for_subclass():