Add __len__ method to IndexBatchSamplerWrapper (#7681)

kaushikb11 · pre-commit-ci[bot] · Borda · commit 78a4f0de5209 · 2021-06-01T12:23:29.000+02:00
Co-authored-by: pre-commit-ci[bot] &lt;66853113+pre-commit-ci[bot]@users.noreply.github.com&gt;
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -5,20 +5,200 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 
 
-## [1.3.3] - 2021-05-27
+## [1.4.0] - 2021-MM-DD
+
+### Added
+
+- Added support to `LightningModule.to_torchscript` for saving to custom filesystems with fsspec ([#7617](https://github.com/PyTorchLightning/pytorch-lightning/pull/7617))
+
+
+- Added `KubeflowEnvironment` for use with the `PyTorchJob` operator in Kubeflow
+
+
+- Added LightningCLI support for config files on object stores ([#7521](https://github.com/PyTorchLightning/pytorch-lightning/pull/7521))
+
+
+- Added `ModelPruning(prune_on_train_epoch_end=True|False)` to choose when to apply pruning ([#7704](https://github.com/PyTorchLightning/pytorch-lightning/pull/7704))
+
+
+- Added support for checkpointing based on a provided time interval during training ([#7515](https://github.com/PyTorchLightning/pytorch-lightning/pull/7515))
+
+
+- Added dataclasses for progress tracking (
+    [#6603](https://github.com/PyTorchLightning/pytorch-lightning/pull/6603),
+    [#7574](https://github.com/PyTorchLightning/pytorch-lightning/pull/7574))
+
+
+- Added argument `trainer.predict(ckpt_path)` ([#7430](https://github.com/PyTorchLightning/pytorch-lightning/pull/7430))
+
+
+- Added `clip_grad_by_value` support for TPUs ([#7025](https://github.com/PyTorchLightning/pytorch-lightning/pull/7025))
+
+
+- Added `sub_dir` parameter to `TensorBoardLogger` ([#6195](https://github.com/PyTorchLightning/pytorch-lightning/pull/6195))
+
+
+- Added correct `dataloader_idx` to batch transfer hooks ([#6241](https://github.com/PyTorchLightning/pytorch-lightning/pull/6241))
+
+
+- Added `ddp_fully_sharded` support ([#7487](https://github.com/PyTorchLightning/pytorch-lightning/pull/7487))
+
+
+- Added `__len__` to `IndexBatchSamplerWrapper` ([#7681](https://github.com/PyTorchLightning/pytorch-lightning/pull/7681))
+
+
+- Added `should_rank_save_checkpoint` property to Training Plugins ([#7684](https://github.com/PyTorchLightning/pytorch-lightning/pull/7684))
+
+
+- Added `save_config_filename` init argument to `LightningCLI` to ease resolving name conflicts ([#7741](https://github.com/PyTorchLightning/pytorch-lightning/pull/7741))
+
 
 ### Changed
 
-- Changed calling of `untoggle_optimizer(opt_idx)` out of the closure function ([#7563](https://github.com/PyTorchLightning/pytorch-lightning/pull/7563))
+- Changed calling of `untoggle_optimizer(opt_idx)` out of the closure function ([#7563](https://github.com/PyTorchLightning/pytorch-lightning/pull/7563)
+
+- Changed the `Trainer`'s `checkpoint_callback` argument to allow only boolean values ([#7539](https://github.com/PyTorchLightning/pytorch-lightning/pull/7539))
+
+
+- Log epoch metrics before the `on_evaluation_end` hook ([#7272](https://github.com/PyTorchLightning/pytorch-lightning/pull/7272))
+
+
+- Changed these `Trainer` methods to be protected: `call_setup_hook`, `call_configure_sharded_model`, `pre_dispatch`, `dispatch`, `post_dispatch`, `call_teardown_hook`, `run_train`, `run_sanity_check`, `run_evaluate`, `run_evaluation`, `run_predict`, `track_output_for_epoch_end`
+
+
+- Changed `clip_grad_norm` to use `torch.nn.utils.clip_grad_norm_` ([#7025](https://github.com/PyTorchLightning/pytorch-lightning/pull/7025))
+
+
+- Validation is now always run inside the training epoch scope ([#7357](https://github.com/PyTorchLightning/pytorch-lightning/pull/7357))
+
+
+- Refactored Loops
+    * Moved attributes `global_step`, `current_epoch`, `max/min_steps`, `max/min_epochs`, `batch_idx`, and `total_batch_idx` to TrainLoop ([#7437](https://github.com/PyTorchLightning/pytorch-lightning/pull/7025))
+    * Refactored result handling in training loop ([#7506](https://github.com/PyTorchLightning/pytorch-lightning/pull/7506))
+    * Moved attributes `hiddens` and `split_idx` to TrainLoop ([#7507](https://github.com/PyTorchLightning/pytorch-lightning/pull/7507))
+    * Refactored the logic around manual and automatic optimization inside the optimizer loop ([#7526](https://github.com/PyTorchLightning/pytorch-lightning/pull/7526))
+    * Simplified "should run validation" logic ([#7682](https://github.com/PyTorchLightning/pytorch-lightning/pull/7682))
+    * Refactored "should run validation" logic when the trainer is signaled to stop ([#7701](https://github.com/PyTorchLightning/pytorch-lightning/pull/7701))
+
+
+- Refactored logging
+    * Renamed and moved `core/step_result.py` to `trainer/connectors/logger_connector/result.py` ([#7736](https://github.com/PyTorchLightning/pytorch-lightning/pull/7736))
+
+
+- Moved `ignore_scalar_return_in_dp` warning suppression to the DataParallelPlugin class ([#7421](https://github.com/PyTorchLightning/pytorch-lightning/pull/7421/))
+
+
+- Changed the behaviour when logging evaluation step metrics to no longer append `/epoch_*` to the metric name ([#7351](https://github.com/PyTorchLightning/pytorch-lightning/pull/7351))
+
+
+- Changed `resolve_training_type_plugins` to allow setting `num_nodes` and `sync_batchnorm` from `Trainer` setting ([#7026](https://github.com/PyTorchLightning/pytorch-lightning/pull/7026))
+
+
+- Default `seed_everything(workers=True)` in the `LightningCLI` ([#7504](https://github.com/PyTorchLightning/pytorch-lightning/pull/7504))
+
+
+- Changed `model.state_dict()` in `CheckpointConnector` to allow `training_type_plugin` to customize the model's `state_dict()` ([#7474](https://github.com/PyTorchLightning/pytorch-lightning/pull/7474))
+
+
+- MLflowLogger now uses the env variable `MLFLOW_TRACKING_URI` as default tracking uri ([#7457](https://github.com/PyTorchLightning/pytorch-lightning/pull/7457))
+
+
+- Changed `WandbLogger(log_model={True/'all'})` to log models as artifacts ([#6231](https://github.com/PyTorchLightning/pytorch-lightning/pull/6231))
+- MLFlowLogger now accepts `run_name` as an constructor argument ([#7622](https://github.com/PyTorchLightning/pytorch-lightning/issues/7622))
+
+
+- Changed `teardown()` in `Accelerator` to allow `training_type_plugin` to customize `teardown` logic ([#7579](https://github.com/PyTorchLightning/pytorch-lightning/pull/7579))
+
+
+### Deprecated
+
+
+- Deprecated `TrainerModelHooksMixin` in favor of `pytorch_lightning.utilities.signature_utils` ([#7422](https://github.com/PyTorchLightning/pytorch-lightning/pull/7422))
+
+
+- Deprecated `num_nodes` and `sync_batchnorm` arguments in `DDPPlugin` and `DDPSpawnPlugin` ([#7026](https://github.com/PyTorchLightning/pytorch-lightning/pull/7026))
+
+
+### Removed
+
+- Removed `ProfilerConnector` ([#7654](https://github.com/PyTorchLightning/pytorch-lightning/pull/7654))
+
+
+- Prune deprecated classif. metrics from `pytorch_lightning.metrics.functional.classification` ([#7499](https://github.com/PyTorchLightning/pytorch-lightning/pull/7499))
+
+
+- Removed deprecated data parallel classes `LightningDataParallel` and `LightningDistributedDataParallel` from `pytorch_lightning.overrides.data_parallel` ([#7510](https://github.com/PyTorchLightning/pytorch-lightning/pull/7510))
+
+
+- Removed deprecated trainer attributes - `get_model` and `accelerator_backend` ([#7502](https://github.com/PyTorchLightning/pytorch-lightning/pull/7502))
+
+
+- Removed support for `self.log(tbptt_reduce_fx)` and `self.log(tbptt_pad_token)`. Please, open a discussion explaining your use-case if you relied on these. ([#7644](https://github.com/PyTorchLightning/pytorch-lightning/pull/7644))
+
+
+- Removed deprecated utils modules `model_utils`, `warning_utils`, `xla_device_utils` and partially `argparse_utils` ([#7503](https://github.com/PyTorchLightning/pytorch-lightning/pull/7503))
+
+
+- Removed deprecated trainer attributes - `on_cpu`, `on_tpu`, `use_tpu`, `on_gpu`, `use_dp`, `use_ddp`, `use_ddp2`, `use_horovod`, `use_single_gpu` ([#7501](https://github.com/PyTorchLightning/pytorch-lightning/pull/7501))
+
 
 ### Fixed
 
-- Fixed `ProgressBar` pickling after calling `trainer.predict` ([#7608](https://github.com/PyTorchLightning/pytorch-lightning/pull/7608))
-- Fixed broadcasting in multi-node, multi-gpu DDP using torch 1.7 ([#7592](https://github.com/PyTorchLightning/pytorch-lightning/pull/7592))
+- Fixed ambiguous warning when both overfit and train dataloader shuffling are enabled ([#7685](https://github.com/PyTorchLightning/pytorch-lightning/pull/7685))
+
 - Fixed dataloaders are not reset when tuning the model ([#7566](https://github.com/PyTorchLightning/pytorch-lightning/pull/7566))
-- Fixed print errors in `ProgressBar` when `trainer.fit` is not called ([#7674](https://github.com/PyTorchLightning/pytorch-lightning/pull/7674))
+
+
 - Fixed global step update when the epoch is skipped ([#7677](https://github.com/PyTorchLightning/pytorch-lightning/pull/7677))
+
+
 - Fixed training loop total batch counter when accumulate grad batches was enabled ([#7692](https://github.com/PyTorchLightning/pytorch-lightning/pull/7692))
+
+
+- Fixed broadcasting in multi-node, multi-gpu DDP using torch 1.7 ([#7592](https://github.com/PyTorchLightning/pytorch-lightning/pull/7592))
+
+
+- Fixed `ProgressBar` pickling after calling `trainer.predict` ([#7608](https://github.com/PyTorchLightning/pytorch-lightning/pull/7608))
+
+
+- Fixed `None` loss keys getting added in `training_epoch_end` when using manual optimization and not returning a loss ([#7772](https://github.com/PyTorchLightning/pytorch-lightning/pull/7772))
+
+
+- Fixed print errors in `ProgressBar` when `trainer.fit` is not called ([#7674](https://github.com/PyTorchLightning/pytorch-lightning/pull/7674))
+
+
+- Fixed formatting of info message when max training time reached ([#7780](https://github.com/PyTorchLightning/pytorch-lightning/pull/7780))
+
+
+## [1.3.4] - 2021-06-01
+
+### Changed
+
+### Fixed
+
+- Update pre-commit and add new hooks ([#7781](https://github.com/PyTorchLightning/pytorch-lightning/pull/7781))
+- fix info message when max training time reached ([#7780](https://github.com/PyTorchLightning/pytorch-lightning/pull/7780))
+- Fixed missing __len__ method to IndexBatchSamplerWrapper ([#7681](https://github.com/PyTorchLightning/pytorch-lightning/pull/7681))
+
+
+
+
+## [1.3.3] - 2021-05-27
+
+### Changed
+
+
+- Move parameter validation specific to TPU Training plugins ([#7415](https://github.com/PyTorchLightning/pytorch-lightning/pull/7415))
+
+### Fixed
+
+- Fixed broadcasting in multi-node, multi-gpu DDP using torch 1.7 ([#7592](https://github.com/PyTorchLightning/pytorch-lightning/pull/7592))
+- Fixed training loop total batch counter when accumulate grad batches was enabled ([#7692](https://github.com/PyTorchLightning/pytorch-lightning/pull/7692))
+- Fixed global step update when the epoch is skipped ([#7677](https://github.com/PyTorchLightning/pytorch-lightning/pull/7677))
+- Fixed print errors in `ProgressBar` when `trainer.fit` is not called ([#7674](https://github.com/PyTorchLightning/pytorch-lightning/pull/7674))
+- Fixed dataloaders are not reset when tuning the model ([#7566](https://github.com/PyTorchLightning/pytorch-lightning/pull/7566))
+- Fix/mismatched toggle optimizer ([#7563](https://github.com/PyTorchLightning/pytorch-lightning/pull/7563))
+- Fixed `ProgressBar` pickling after calling `trainer.predict` ([#7608](https://github.com/PyTorchLightning/pytorch-lightning/pull/7608))
 - Fixed formatting of info message when max training time reached ([#7780](https://github.com/PyTorchLightning/pytorch-lightning/pull/7780))
 
 
diff --git a/pytorch_lightning/overrides/distributed.py b/pytorch_lightning/overrides/distributed.py
@@ -132,6 +132,9 @@ def __iter__(self) -> Iterator[List[int]]:
             self.batch_indices = batch
             yield batch
 
+    def __len__(self) -> int:
+        return len(self._sampler)
+
     @property
     def drop_last(self) -> bool:
         return self._sampler.drop_last
diff --git a/tests/overrides/test_distributed.py b/tests/overrides/test_distributed.py
@@ -11,11 +11,14 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
+from collections.abc import Iterable
+
 import pytest
 from torch.utils.data import BatchSampler, SequentialSampler
 
 from pytorch_lightning import seed_everything
 from pytorch_lightning.overrides.distributed import IndexBatchSamplerWrapper, UnrepeatedDistributedSampler
+from pytorch_lightning.utilities.data import has_len
 
 
 @pytest.mark.parametrize("shuffle", [False, True])
@@ -54,3 +57,13 @@ def test_index_batch_sampler(tmpdir):
 
     for batch in index_batch_sampler:
         assert index_batch_sampler.batch_indices == batch
+
+
+def test_index_batch_sampler_methods():
+    dataset = range(15)
+    sampler = SequentialSampler(dataset)
+    batch_sampler = BatchSampler(sampler, 3, False)
+    index_batch_sampler = IndexBatchSamplerWrapper(batch_sampler)
+
+    assert isinstance(index_batch_sampler, Iterable)
+    assert has_len(index_batch_sampler)
