Merge branch 'master' into bugfix/should_stop

Author: awaelchli

CHANGELOG.md

@@ -170,6 +170,9 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 
 ### Fixed
 
+- Sanitize `None` params during pruning ([#6836](https://github.com/PyTorchLightning/pytorch-lightning/pull/6836))
+
+
 - Made the `Plugin.reduce` method more consistent across all Plugins to reflect a mean-reduction by default ([#6011](https://github.com/PyTorchLightning/pytorch-lightning/pull/6011))
 
 
@@ -197,9 +200,16 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 - Fixed torch distributed not available in setup hook for DDP ([#6506](https://github.com/PyTorchLightning/pytorch-lightning/pull/6506))
 
 
+- Enforce an epoch scheduler interval when using SWA ([#6588](https://github.com/PyTorchLightning/pytorch-lightning/pull/6588))
+
+
+- Fixed an issue with `IterableDataset` when `__len__` is not defined ([#6828](https://github.com/PyTorchLightning/pytorch-lightning/pull/6828))
+
+
 - Fixed `EarlyStopping` logic when `min_epochs` or `min_steps` requirement is not met ([#6705](https://github.com/PyTorchLightning/pytorch-lightning/pull/6705))
 
 
+
 ## [1.2.6] - 2021-03-30
 
 ### Changed

docs/source/advanced/tpu.rst

@@ -64,8 +64,7 @@ To get a TPU on colab, follow these steps:
 
    .. code-block::
 
-        !curl https://raw.githubusercontent.com/pytorch/xla/master/contrib/scripts/env-setup.py -o pytorch-xla-env-setup.py
-        !python pytorch-xla-env-setup.py --version 1.7 --apt-packages libomp5 libopenblas-dev
+        !pip install cloud-tpu-client==0.10 https://storage.googleapis.com/tpu-pytorch/wheels/torch_xla-1.8-cp37-cp37m-linux_x86_64.whl
 
 5. Once the above is done, install PyTorch Lightning (v 0.7.0+).
 

docs/source/common/lightning_module.rst

@@ -912,30 +912,6 @@ use_amp
 ~~~~~~~
 True if using Automatic Mixed Precision (AMP)
 
-------------
-
-use_ddp
-~~~~~~~
-True if using ddp
-
-------------
-
-use_ddp2
-~~~~~~~~
-True if using ddp2
-
-------------
-
-use_dp
-~~~~~~
-True if using dp
-
-------------
-
-use_tpu
-~~~~~~~
-True if using TPUs
-
 --------------
 
 automatic_optimization

docs/source/starter/introduction_guide.rst

@@ -572,9 +572,7 @@ Next, install the required xla library (adds support for PyTorch on TPUs)
 
 .. code-block:: shell
 
-    !curl https://raw.githubusercontent.com/pytorch/xla/master/contrib/scripts/env-setup.py -o pytorch-xla-env-setup.py
-
-    !python pytorch-xla-env-setup.py --version nightly --apt-packages libomp5 libopenblas-dev
+    !pip install cloud-tpu-client==0.10 https://storage.googleapis.com/tpu-pytorch/wheels/torch_xla-1.8-cp37-cp37m-linux_x86_64.whl
 
 In distributed training (multiple GPUs and multiple TPU cores) each GPU or TPU core will run a copy
 of this program. This means that without taking any care you will download the dataset N times which

pytorch_lightning/accelerators/accelerator.py

@@ -480,7 +480,7 @@ def connect_precision_plugin(self, plugin: PrecisionPlugin) -> None:
         )
         self.setup_precision_plugin(plugin)
 
-    def save_checkpoint(self, checkpoint: Dict[str, Any], filepath) -> None:
+    def save_checkpoint(self, checkpoint: Dict[str, Any], filepath: str) -> None:
         """Save model/training states as a checkpoint file through state-dump and file-write.
 
         Args:

pytorch_lightning/callbacks/finetuning.py

@@ -77,7 +77,7 @@ def finetune_function(self, pl_module, current_epoch, optimizer, optimizer_idx):
                 # When `current_epoch` is 10, feature_extractor will start training.
                 if current_epoch == self._unfreeze_at_epoch:
                     self.unfreeze_and_add_param_group(
-                        module=pl_module.feature_extractor,
+                        modules=pl_module.feature_extractor,
                         optimizer=optimizer,
                         train_bn=True,
                     )

pytorch_lightning/callbacks/progress.py

@@ -148,9 +148,10 @@ def total_val_batches(self) -> int:
         validation dataloader is of infinite size.
         """
         total_val_batches = 0
-        if not self.trainer.disable_validation:
-            is_val_epoch = (self.trainer.current_epoch) % self.trainer.check_val_every_n_epoch == 0
+        if self.trainer.enable_validation:
+            is_val_epoch = (self.trainer.current_epoch + 1) % self.trainer.check_val_every_n_epoch == 0
             total_val_batches = sum(self.trainer.num_val_batches) if is_val_epoch else 0
+
         return total_val_batches
 
     @property

pytorch_lightning/callbacks/pruning.py

@@ -422,7 +422,9 @@ def sanitize_parameters_to_prune(
         current_modules = [m for m in pl_module.modules() if not isinstance(m, _MODULE_CONTAINERS)]
 
         if parameters_to_prune is None:
-            parameters_to_prune = [(m, p) for p in parameters for m in current_modules if hasattr(m, p)]
+            parameters_to_prune = [
+                (m, p) for p in parameters for m in current_modules if getattr(m, p, None) is not None
+            ]
         elif (
             isinstance(parameters_to_prune, (list, tuple)) and len(parameters_to_prune) > 0
             and all(len(p) == 2 for p in parameters_to_prune)

pytorch_lightning/callbacks/stochastic_weight_avg.py

@@ -187,14 +187,15 @@ def on_train_epoch_start(self, trainer: 'pl.Trainer', pl_module: 'pl.LightningMo
                 anneal_strategy=self._annealing_strategy,
                 last_epoch=trainer.max_epochs if self._annealing_strategy == "cos" else -1
             )
+            _scheduler_config = _get_default_scheduler_config()
+            assert _scheduler_config["interval"] == "epoch" and _scheduler_config["frequency"] == 1
+            _scheduler_config["scheduler"] = self._swa_scheduler
 
             if trainer.lr_schedulers:
                 lr_scheduler = trainer.lr_schedulers[0]["scheduler"]
                 rank_zero_warn(f"Swapping lr_scheduler {lr_scheduler} for {self._swa_scheduler}")
-                trainer.lr_schedulers[0]["scheduler"] = self._swa_scheduler
+                trainer.lr_schedulers[0] = _scheduler_config
             else:
-                _scheduler_config = _get_default_scheduler_config()
-                _scheduler_config["scheduler"] = self._swa_scheduler
                 trainer.lr_schedulers.append(_scheduler_config)
 
             self.n_averaged = torch.tensor(0, dtype=torch.long, device=pl_module.device)

pytorch_lightning/metrics/classification/accuracy.py

@@ -37,3 +37,4 @@ def __init__(
         .. deprecated::
             Use :class:`~torchmetrics.Accuracy`. Will be removed in v1.5.0.
         """
+        _ = threshold, top_k, subset_accuracy, compute_on_step, dist_sync_on_step, process_group, dist_sync_fn