Rename flag to_review -> under_review (#837)

docs/source/stability.rst

@@ -19,8 +19,8 @@ For stable features, all of the following are true:
 - if anything does change, incorrect usage will give a deprecation warning for **one minor release** before the breaking change is made
 - the API has been tested for compatibility with latest releases of PyTorch Lightning and Flash
 
-To Review
-_________
+Under Review
+____________
 
 For features to be reviewed, any or all of the following may be true:
 

pl_bolts/callbacks/byol_updates.py

@@ -5,10 +5,10 @@
 from torch import Tensor
 from torch.nn import Module
 
-from pl_bolts.utils.stability import to_review
+from pl_bolts.utils.stability import under_review
 
 
-@to_review()
+@under_review()
 class BYOLMAWeightUpdate(Callback):
     """Weight update rule from BYOL.
 

pl_bolts/callbacks/data_monitor.py

@@ -11,7 +11,7 @@
 from torch.utils.hooks import RemovableHandle
 
 from pl_bolts.utils import _WANDB_AVAILABLE
-from pl_bolts.utils.stability import to_review
+from pl_bolts.utils.stability import under_review
 from pl_bolts.utils.warnings import warn_missing_pkg
 
 if _WANDB_AVAILABLE:
@@ -20,7 +20,7 @@
     warn_missing_pkg("wandb")
 
 
-@to_review()
+@under_review()
 class DataMonitorBase(Callback):
 
     supported_loggers = (
@@ -111,7 +111,7 @@ def _is_logger_available(self, logger: LightningLoggerBase) -> bool:
         return available
 
 
-@to_review()
+@under_review()
 class ModuleDataMonitor(DataMonitorBase):
 
     GROUP_NAME_INPUT = "input"
@@ -197,7 +197,7 @@ def hook(_: Module, inp: Sequence, out: Sequence) -> None:
         return handle
 
 
-@to_review()
+@under_review()
 class TrainingDataMonitor(DataMonitorBase):
 
     GROUP_NAME = "training_step"
@@ -261,7 +261,7 @@ def collect_and_name_tensors(data: Any, output: Dict[str, Tensor], parent_name:
             collect_and_name_tensors(item, output, parent_name=f"{parent_name}/{i:d}")
 
 
-@to_review()
+@under_review()
 def shape2str(tensor: Tensor) -> str:
     """Returns the shape of a tensor in bracket notation as a string.
 

pl_bolts/callbacks/knn_online.py

@@ -6,10 +6,10 @@
 from torch import Tensor
 from torch.nn import functional as F
 
-from pl_bolts.utils.stability import to_review
+from pl_bolts.utils.stability import under_review
 
 
-@to_review()
+@under_review()
 class KNNOnlineEvaluator(Callback):
     """Weighted KNN online evaluator for self-supervised learning.
     The weighted KNN classifier matches sec 3.4 of https://arxiv.org/pdf/1805.01978.pdf.
@@ -141,6 +141,6 @@ def on_validation_epoch_end(self, trainer: Trainer, pl_module: LightningModule)
         pl_module.log("online_knn_val_acc", total_top1 / total_num, on_step=False, on_epoch=True, sync_dist=True)
 
 
-@to_review()
+@under_review()
 def concat_all_gather(tensor: Tensor, accelerator: Accelerator) -> Tensor:
     return accelerator.all_gather(tensor).view(-1, *tensor.shape[1:])

pl_bolts/callbacks/printing.py

@@ -6,10 +6,10 @@
 from pytorch_lightning.callbacks import Callback
 from pytorch_lightning.utilities import rank_zero_info
 
-from pl_bolts.utils.stability import to_review
+from pl_bolts.utils.stability import under_review
 
 
-@to_review()
+@under_review()
 class PrintTableMetricsCallback(Callback):
     """Prints a table with the metrics in columns on every epoch end.
 
@@ -44,7 +44,7 @@ def on_epoch_end(self, trainer: Trainer, pl_module: LightningModule) -> None:
         rank_zero_info(dicts_to_table(self.metrics))
 
 
-@to_review()
+@under_review()
 def dicts_to_table(
     dicts: List[Dict],
     keys: Optional[List[str]] = None,

pl_bolts/callbacks/sparseml.py

@@ -23,10 +23,10 @@
     from sparseml.pytorch.optim import ScheduledModifierManager
     from sparseml.pytorch.utils import ModuleExporter
 
-from pl_bolts.utils.stability import to_review
+from pl_bolts.utils.stability import under_review
 
 
-@to_review()
+@under_review()
 class SparseMLCallback(Callback):
     """Enables SparseML aware training. Requires a recipe to run during training.
 

pl_bolts/callbacks/ssl_online.py

@@ -10,10 +10,10 @@
 from torchmetrics.functional import accuracy
 
 from pl_bolts.models.self_supervised.evaluator import SSLEvaluator
-from pl_bolts.utils.stability import to_review
+from pl_bolts.utils.stability import under_review
 
 
-@to_review()
+@under_review()
 class SSLOnlineEvaluator(Callback):  # pragma: no cover
     """Attaches a MLP for fine-tuning using the standard self-supervised protocol.
 
@@ -175,7 +175,7 @@ def on_load_checkpoint(self, trainer: Trainer, pl_module: LightningModule, callb
         self._recovered_callback_state = callback_state
 
 
-@to_review()
+@under_review()
 @contextmanager
 def set_training(module: nn.Module, mode: bool):
     """Context manager to set training mode.

pl_bolts/callbacks/torch_ort.py

@@ -19,10 +19,10 @@
 if _TORCH_ORT_AVAILABLE:
     from torch_ort import ORTModule
 
-from pl_bolts.utils.stability import to_review
+from pl_bolts.utils.stability import under_review
 
 
-@to_review()
+@under_review()
 class ORTCallback(Callback):
     """Enables Torch ORT: Accelerate PyTorch models with ONNX Runtime.
 

pl_bolts/callbacks/variational.py

@@ -7,7 +7,7 @@
 from torch import Tensor
 
 from pl_bolts.utils import _TORCHVISION_AVAILABLE
-from pl_bolts.utils.stability import to_review
+from pl_bolts.utils.stability import under_review
 from pl_bolts.utils.warnings import warn_missing_pkg
 
 if _TORCHVISION_AVAILABLE:
@@ -16,7 +16,7 @@
     warn_missing_pkg("torchvision")
 
 
-@to_review()
+@under_review()
 class LatentDimInterpolator(Callback):
     """Interpolates the latent space for a model by setting all dims to zero and stepping through the first two
     dims increasing one unit at a time.

pl_bolts/callbacks/verification/base.py

@@ -9,10 +9,10 @@
 from pytorch_lightning.utilities import move_data_to_device, rank_zero_warn
 from pytorch_lightning.utilities.signature_utils import is_param_in_hook_signature
 
-from pl_bolts.utils.stability import to_review
+from pl_bolts.utils.stability import under_review
 
 
-@to_review()
+@under_review()
 class VerificationBase:
     """Base class for model verification.
 
@@ -82,7 +82,7 @@ def _model_forward(self, input_array: Any) -> Any:
         return self.model(input_array)
 
 
-@to_review()
+@under_review()
 class VerificationCallbackBase(Callback):
     """Base class for model verification in form of a callback.
 

pl_bolts/callbacks/verification/batch_gradient.py

@@ -10,10 +10,10 @@
 from torch import Tensor
 
 from pl_bolts.callbacks.verification.base import VerificationBase, VerificationCallbackBase
-from pl_bolts.utils.stability import to_review
+from pl_bolts.utils.stability import under_review
 
 
-@to_review()
+@under_review()
 class BatchGradientVerification(VerificationBase):
     """Checks if a model mixes data across the batch dimension.
 
@@ -84,7 +84,7 @@ def check(
         return not any(has_grad_outside_sample) and all(has_grad_inside_sample)
 
 
-@to_review()
+@under_review()
 class BatchGradientVerificationCallback(VerificationCallbackBase):
     """The callback version of the :class:`BatchGradientVerification` test.
 
@@ -133,7 +133,7 @@ def on_train_start(self, trainer: Trainer, pl_module: LightningModule) -> None:
             self._raise()
 
 
-@to_review()
+@under_review()
 def default_input_mapping(data: Any) -> List[Tensor]:
     """Finds all tensors in a (nested) collection that have the same batch size.
 
@@ -161,7 +161,7 @@ def default_input_mapping(data: Any) -> List[Tensor]:
     return batches
 
 
-@to_review()
+@under_review()
 def default_output_mapping(data: Any) -> Tensor:
     """Pulls out all tensors in a output collection and combines them into one big batch for verification.
 
@@ -193,7 +193,7 @@ def default_output_mapping(data: Any) -> Tensor:
     return combined
 
 
-@to_review()
+@under_review()
 def collect_tensors(data: Any) -> List[Tensor]:
     """Filters all tensors in a collection and returns them in a list."""
     tensors = []
@@ -206,7 +206,7 @@ def collect_batches(tensor: Tensor) -> Tensor:
     return tensors
 
 
-@to_review()
+@under_review()
 @contextmanager
 def selective_eval(model: nn.Module, layer_types: Iterable[Type[nn.Module]]) -> None:
     """A context manager that sets all requested types of layers to eval mode. This method uses an ``isinstance``

pl_bolts/callbacks/vision/confused_logit.py

@@ -5,7 +5,7 @@
 from torch import Tensor, nn
 
 from pl_bolts.utils import _MATPLOTLIB_AVAILABLE
-from pl_bolts.utils.stability import to_review
+from pl_bolts.utils.stability import under_review
 from pl_bolts.utils.warnings import warn_missing_pkg
 
 if _MATPLOTLIB_AVAILABLE:
@@ -18,7 +18,7 @@
     Figure = object
 
 
-@to_review()
+@under_review()
 class ConfusedLogitCallback(Callback):  # pragma: no cover
     """Takes the logit predictions of a model and when the probabilities of two classes are very close, the model
     doesn't have high certainty that it should pick one vs the other class.

pl_bolts/callbacks/vision/image_generation.py

@@ -4,7 +4,7 @@
 from pytorch_lightning import Callback, LightningModule, Trainer
 
 from pl_bolts.utils import _TORCHVISION_AVAILABLE
-from pl_bolts.utils.stability import to_review
+from pl_bolts.utils.stability import under_review
 from pl_bolts.utils.warnings import warn_missing_pkg
 
 if _TORCHVISION_AVAILABLE:
@@ -13,7 +13,7 @@
     warn_missing_pkg("torchvision")
 
 
-@to_review()
+@under_review()
 class TensorboardGenerativeModelImageSampler(Callback):
     """Generates images and logs to tensorboard. Your model must implement the ``forward`` function for generation.
 

pl_bolts/callbacks/vision/sr_image_logger.py

@@ -6,7 +6,7 @@
 from pytorch_lightning import Callback
 
 from pl_bolts.utils import _TORCHVISION_AVAILABLE
-from pl_bolts.utils.stability import to_review
+from pl_bolts.utils.stability import under_review
 from pl_bolts.utils.warnings import warn_missing_pkg
 
 if _TORCHVISION_AVAILABLE:
@@ -15,7 +15,7 @@
     warn_missing_pkg("torchvision")
 
 
-@to_review()
+@under_review()
 class SRImageLoggerCallback(Callback):
     """Logs low-res, generated high-res, and ground truth high-res images to TensorBoard Your model must implement
     the ``forward`` function for generation.

pl_bolts/datamodules/async_dataloader.py

@@ -9,10 +9,10 @@
 from torch._six import string_classes
 from torch.utils.data import DataLoader, Dataset
 
-from pl_bolts.utils.stability import to_review
+from pl_bolts.utils.stability import under_review
 
 
-@to_review()
+@under_review()
 class AsynchronousLoader:
     """Class for asynchronously loading from CPU memory to device memory with DataLoader.
 

pl_bolts/datamodules/binary_emnist_datamodule.py

@@ -3,10 +3,10 @@
 from pl_bolts.datamodules.emnist_datamodule import EMNISTDataModule
 from pl_bolts.datasets import BinaryEMNIST
 from pl_bolts.utils import _TORCHVISION_AVAILABLE
-from pl_bolts.utils.stability import to_review
+from pl_bolts.utils.stability import under_review
 
 
-@to_review()
+@under_review()
 class BinaryEMNISTDataModule(EMNISTDataModule):
     """
     .. figure:: https://user-images.githubusercontent.com/4632336/123210742-4d6b3380-d477-11eb-80da-3e9a74a18a07.png

pl_bolts/datamodules/binary_mnist_datamodule.py

@@ -3,7 +3,7 @@
 from pl_bolts.datamodules.vision_datamodule import VisionDataModule
 from pl_bolts.datasets import BinaryMNIST
 from pl_bolts.utils import _TORCHVISION_AVAILABLE
-from pl_bolts.utils.stability import to_review
+from pl_bolts.utils.stability import under_review
 from pl_bolts.utils.warnings import warn_missing_pkg
 
 if _TORCHVISION_AVAILABLE:
@@ -12,7 +12,7 @@
     warn_missing_pkg("torchvision")
 
 
-@to_review()
+@under_review()
 class BinaryMNISTDataModule(VisionDataModule):
     """
     .. figure:: https://miro.medium.com/max/744/1*AO2rIhzRYzFVQlFLx9DM9A.png

pl_bolts/datamodules/cifar10_datamodule.py

@@ -4,7 +4,7 @@
 from pl_bolts.datasets import TrialCIFAR10
 from pl_bolts.transforms.dataset_normalizations import cifar10_normalization
 from pl_bolts.utils import _TORCHVISION_AVAILABLE
-from pl_bolts.utils.stability import to_review
+from pl_bolts.utils.stability import under_review
 from pl_bolts.utils.warnings import warn_missing_pkg
 
 if _TORCHVISION_AVAILABLE:
@@ -15,7 +15,7 @@
     CIFAR10 = None
 
 
-@to_review()
+@under_review()
 class CIFAR10DataModule(VisionDataModule):
     """
     .. figure:: https://3qeqpr26caki16dnhd19sv6by6v-wpengine.netdna-ssl.com/wp-content/uploads/2019/01/
@@ -124,7 +124,7 @@ def default_transforms(self) -> Callable:
         return cf10_transforms
 
 
-@to_review()
+@under_review()
 class TinyCIFAR10DataModule(CIFAR10DataModule):
     """Standard CIFAR10, train, val, test splits and transforms.