From ca39eb50bd55215491291466a3558f84f180d9ed Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Mon, 26 Apr 2021 13:06:36 +0200
Subject: [PATCH 01/51] Add typing for apply_func

---
 pytorch_lightning/utilities/apply_func.py | 30 +++++++++++++----------
 1 file changed, 17 insertions(+), 13 deletions(-)

diff --git a/pytorch_lightning/utilities/apply_func.py b/pytorch_lightning/utilities/apply_func.py
index e100a803bcd00..d384c0edbd3bf 100644
--- a/pytorch_lightning/utilities/apply_func.py
+++ b/pytorch_lightning/utilities/apply_func.py
@@ -16,7 +16,7 @@
 from collections.abc import Mapping, Sequence
 from copy import copy
 from functools import partial
-from typing import Any, Callable, Optional, Union
+from typing import Any, Callable, List, Optional, Tuple, Union
 
 import numpy as np
 import torch
@@ -33,13 +33,17 @@
     Batch = type(None)
 
 
-def to_dtype_tensor(value, dtype: torch.dtype = None, device: torch.device = None):
+def to_dtype_tensor(
+    value: Union[int, float, List[Union[int, float]]],
+    dtype: Optional[torch.dtype] = None,
+    device: Union[str, torch.device] = None
+) -> torch.Tensor:
     if device is None:
         raise MisconfigurationException("device (torch.device) should be provided.")
     return torch.tensor(value, dtype=dtype, device=device)
 
 
-def from_numpy(value, device: torch.device = None):
+def from_numpy(value: np.ndarray, device: Union[str, torch.device] = None) -> torch.Tensor:
     if device is None:
         raise MisconfigurationException("device (torch.device) should be provided.")
     return torch.from_numpy(value).to(device)
@@ -56,11 +60,11 @@ def from_numpy(value, device: torch.device = None):
 
 def apply_to_collection(
     data: Any,
-    dtype: Union[type, tuple],
+    dtype: Union[torch.dtype, Tuple[torch.dtype]],
     function: Callable,
-    *args,
-    wrong_dtype: Optional[Union[type, tuple]] = None,
-    **kwargs
+    *args: Any,
+    wrong_dtype: Optional[Union[torch.dtype, Tuple[torch.dtype]]] = None,
+    **kwargs: Any
 ) -> Any:
     """
     Recursively applies a function to all elements of a certain dtype.
@@ -87,7 +91,7 @@ def apply_to_collection(
     if isinstance(data, Mapping):
         return elem_type({k: apply_to_collection(v, dtype, function, *args, **kwargs) for k, v in data.items()})
 
-    if isinstance(data, tuple) and hasattr(data, '_fields'):  # named tuple
+    if isinstance(data, Tuple) and hasattr(data, '_fields'):  # named tuple
         return elem_type(*(apply_to_collection(d, dtype, function, *args, **kwargs) for d in data))
 
     if isinstance(data, Sequence) and not isinstance(data, str):
@@ -116,14 +120,14 @@ class TransferableDataType(ABC):
     """
 
     @classmethod
-    def __subclasshook__(cls, subclass):
+    def __subclasshook__(cls, subclass: Any) -> Union[bool, type(NotImplemented)]:
         if cls is TransferableDataType:
             to = getattr(subclass, "to", None)
             return callable(to)
         return NotImplemented
 
 
-def move_data_to_device(batch: Any, device: torch.device):
+def move_data_to_device(batch: Any, device: Union[str, torch.device]) -> Any:
     """
     Transfers a collection of data to the given device. Any object that defines a method
     ``to(device)`` will be moved and all other objects in the collection will be left untouched.
@@ -141,7 +145,7 @@ def move_data_to_device(batch: Any, device: torch.device):
         - :class:`torch.device`
     """
 
-    def batch_to(data):
+    def batch_to(data: Any) -> Any:
         # try to move torchtext data first
         if _TORCHTEXT_AVAILABLE and isinstance(data, Batch):
 
@@ -161,14 +165,14 @@ def batch_to(data):
     return apply_to_collection(batch, dtype=dtype, function=batch_to)
 
 
-def convert_to_tensors(data, device: torch.device = None):
+def convert_to_tensors(data: Any, device: Union[str, torch.device] = None) -> Any:
     if device is None:
         raise MisconfigurationException("device (torch.device) should be provided.")
 
     for src_dtype, conversion_func in CONVERSION_DTYPES:
         data = apply_to_collection(data, src_dtype, partial(conversion_func, device=device))
 
-    def _move_to_device_and_make_contiguous(t: torch.Tensor, device: torch.device):
+    def _move_to_device_and_make_contiguous(t: torch.Tensor, device: Union[str, torch.device]) -> torch.Tensor:
         return t.to(device).contiguous()
 
     data = apply_to_collection(data, torch.Tensor, partial(_move_to_device_and_make_contiguous, device=device))

From 4146be6cdac5883cc4067429d6fcd3591fef7ac0 Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Mon, 26 Apr 2021 13:13:58 +0200
Subject: [PATCH 02/51] Add typing for argparse'

---
 pytorch_lightning/utilities/argparse.py | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/pytorch_lightning/utilities/argparse.py b/pytorch_lightning/utilities/argparse.py
index dc99b923c6702..4a7f96f4447b3 100644
--- a/pytorch_lightning/utilities/argparse.py
+++ b/pytorch_lightning/utilities/argparse.py
@@ -17,10 +17,11 @@
 from contextlib import suppress
 from typing import Any, Dict, List, Tuple, Union
 
+import pytorch_lightning as pl
 from pytorch_lightning.utilities.parsing import str_to_bool, str_to_bool_or_str
 
 
-def from_argparse_args(cls, args: Union[Namespace, ArgumentParser], **kwargs):
+def from_argparse_args(cls, args: Union[Namespace, ArgumentParser], **kwargs: Any) -> 'pl.Trainer':
     """Create an instance from CLI arguments.
     Eventually use varibles from OS environement which are defined as "PL_<CLASS-NAME>_<CLASS_ARUMENT_NAME>"
 
@@ -134,7 +135,7 @@ def get_init_arguments_and_types(cls) -> List[Tuple[str, Tuple, Any]]:
     return name_type_default
 
 
-def _get_abbrev_qualified_cls_name(cls):
+def _get_abbrev_qualified_cls_name(cls) -> str:
     assert isinstance(cls, type), repr(cls)
     if cls.__module__.startswith("pytorch_lightning."):
         # Abbreviate.
@@ -148,7 +149,7 @@ def add_argparse_args(
     cls,
     parent_parser: ArgumentParser,
     *,
-    use_argument_group=True,
+    use_argument_group: bool = True,
 ) -> ArgumentParser:
     r"""Extends existing argparse by default attributes for ``cls``.
 
@@ -279,21 +280,21 @@ def _parse_args_from_docstring(docstring: str) -> Dict[str, str]:
     return parsed
 
 
-def _gpus_allowed_type(x) -> Union[int, str]:
+def _gpus_allowed_type(x: str) -> Union[int, str]:
     if ',' in x:
         return str(x)
     else:
         return int(x)
 
 
-def _gpus_arg_default(x) -> Union[int, str]:  # pragma: no-cover
+def _gpus_arg_default(x: str) -> None:  # pragma: no-cover
     # unused, but here for backward compatibility with old checkpoints that need to be able to
     # unpickle the function from the checkpoint, as it was not filtered out in versions < 1.2.8
     # see: https://github.com/PyTorchLightning/pytorch-lightning/pull/6898
     pass
 
 
-def _int_or_float_type(x) -> Union[int, float]:
+def _int_or_float_type(x: Union[int, float, str]) -> Union[int, float]:
     if '.' in str(x):
         return float(x)
     else:

From 5582fad548794719ee4c2ef09fc976f8c7b58781 Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Mon, 26 Apr 2021 13:23:04 +0200
Subject: [PATCH 03/51] Add typing for cli

---
 pytorch_lightning/utilities/cli.py | 17 +++++++++--------
 1 file changed, 9 insertions(+), 8 deletions(-)

diff --git a/pytorch_lightning/utilities/cli.py b/pytorch_lightning/utilities/cli.py
index 93729a53db25d..730fe2798ef24 100644
--- a/pytorch_lightning/utilities/cli.py
+++ b/pytorch_lightning/utilities/cli.py
@@ -13,8 +13,9 @@
 # limitations under the License.
 import os
 from argparse import Namespace
-from typing import Any, Dict, Optional, Type, Union
+from typing import Any, Dict, Optional, Union
 
+import pytorch_lightning as pl
 from pytorch_lightning.callbacks import Callback
 from pytorch_lightning.core.datamodule import LightningDataModule
 from pytorch_lightning.core.lightning import LightningModule
@@ -32,7 +33,7 @@
 class LightningArgumentParser(ArgumentParser):
     """Extension of jsonargparse's ArgumentParser for pytorch-lightning"""
 
-    def __init__(self, *args, parse_as_dict: bool = True, **kwargs) -> None:
+    def __init__(self, *args: Any, parse_as_dict: bool = True, **kwargs: Any) -> None:
         """Initialize argument parser that supports configuration file input
 
         For full details of accepted arguments see `ArgumentParser.__init__
@@ -50,7 +51,7 @@ def __init__(self, *args, parse_as_dict: bool = True, **kwargs) -> None:
 
     def add_lightning_class_args(
         self,
-        lightning_class: Union[Type[Trainer], Type[LightningModule], Type[LightningDataModule]],
+        lightning_class: Union['pl.Trainer', 'pl.LightningModule', 'pl.LightningDataModule'],
         nested_key: str,
         subclass_mode: bool = False
     ) -> None:
@@ -92,16 +93,16 @@ class LightningCLI:
 
     def __init__(
         self,
-        model_class: Type[LightningModule],
-        datamodule_class: Type[LightningDataModule] = None,
-        save_config_callback: Type[SaveConfigCallback] = SaveConfigCallback,
-        trainer_class: Type[Trainer] = Trainer,
+        model_class: 'pl.LightningModule',
+        datamodule_class: Optional['pl.LightningDataModule'] = None,
+        save_config_callback: SaveConfigCallback = SaveConfigCallback,
+        trainer_class: 'pl.Trainer' = Trainer,
         trainer_defaults: Dict[str, Any] = None,
         seed_everything_default: int = None,
         description: str = 'pytorch-lightning trainer command line tool',
         env_prefix: str = 'PL',
         env_parse: bool = False,
-        parser_kwargs: Dict[str, Any] = None,
+        parser_kwargs: Optional[Dict[str, Any]] = None,
         subclass_mode_model: bool = False,
         subclass_mode_data: bool = False
     ) -> None:

From 5a432ddd431d8b4d27bc5298f5d1251e4bcdb875 Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Mon, 26 Apr 2021 13:48:54 +0200
Subject: [PATCH 04/51] Add typing for cloud_io

---
 pytorch_lightning/utilities/cloud_io.py | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/pytorch_lightning/utilities/cloud_io.py b/pytorch_lightning/utilities/cloud_io.py
index 8b65ca143976e..07d52d88d88c5 100644
--- a/pytorch_lightning/utilities/cloud_io.py
+++ b/pytorch_lightning/utilities/cloud_io.py
@@ -15,7 +15,7 @@
 import io
 import os
 from pathlib import Path
-from typing import IO, Union
+from typing import Any, Dict, IO, Optional, Union
 
 import fsspec
 import torch
@@ -34,7 +34,10 @@ def isdir(self, path: str) -> bool:
         return os.path.isdir(path)  # follows symlinks
 
 
-def load(path_or_url: Union[str, IO, Path], map_location=None):
+def load(
+    path_or_url: Union[str, IO, Path],
+    map_location: Optional[Union[str, torch.device, Dict[Union[str, torch.device], Union[str, torch.device]]]] = None
+) -> Any:
     if not isinstance(path_or_url, (str, Path)):
         # any sort of BytesIO or similiar
         return torch.load(path_or_url, map_location=map_location)
@@ -45,7 +48,7 @@ def load(path_or_url: Union[str, IO, Path], map_location=None):
         return torch.load(f, map_location=map_location)
 
 
-def get_filesystem(path: Union[str, Path]):
+def get_filesystem(path: Union[str, Path]) -> LocalFileSystem:
     path = str(path)
     if "://" in path:
         # use the fileystem from the protocol specified
@@ -55,7 +58,7 @@ def get_filesystem(path: Union[str, Path]):
         return _LightningLocalFileSystem()
 
 
-def atomic_save(checkpoint, filepath: str):
+def atomic_save(checkpoint: Any, filepath: str) -> None:
     """Saves a checkpoint atomically, avoiding the creation of incomplete checkpoints.
 
     Args:

From 71da6853a47267d6cb2e294c183c4bfa340b8638 Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Mon, 26 Apr 2021 15:18:17 +0200
Subject: [PATCH 05/51] Add typing for data and debugging

* [WIP] Need to verify if logged metrics are of a type torch.Tensor
or whether Any can be passed in
---
 pytorch_lightning/utilities/data.py      |  2 +-
 pytorch_lightning/utilities/debugging.py | 46 +++++++++++++++---------
 2 files changed, 30 insertions(+), 18 deletions(-)

diff --git a/pytorch_lightning/utilities/data.py b/pytorch_lightning/utilities/data.py
index 27345fda3b110..8f22d1668d831 100644
--- a/pytorch_lightning/utilities/data.py
+++ b/pytorch_lightning/utilities/data.py
@@ -19,7 +19,7 @@
 from pytorch_lightning.utilities import rank_zero_warn
 
 
-def has_iterable_dataset(dataloader: DataLoader):
+def has_iterable_dataset(dataloader: DataLoader) -> bool:
     return hasattr(dataloader, 'dataset') and isinstance(dataloader.dataset, IterableDataset)
 
 
diff --git a/pytorch_lightning/utilities/debugging.py b/pytorch_lightning/utilities/debugging.py
index 56833fd03735a..58fb1aeffc2ef 100644
--- a/pytorch_lightning/utilities/debugging.py
+++ b/pytorch_lightning/utilities/debugging.py
@@ -16,10 +16,15 @@
 import time
 from collections import Counter
 from functools import wraps
-from typing import Any, Callable, Optional
+from typing import Any, Dict, Callable, List, Optional
 
+import torch
+from torch.utils.data import DataLoader
 
-def enabled_only(fn: Callable):
+import pytorch_lightning as pl
+
+
+def enabled_only(fn: Callable) -> Optional[Callable]:
     """Decorate a logger method to run it only on the process with rank 0.
 
     Args:
@@ -27,7 +32,7 @@ def enabled_only(fn: Callable):
     """
 
     @wraps(fn)
-    def wrapped_fn(self, *args, **kwargs):
+    def wrapped_fn(self, *args: Any, **kwargs: Any) -> Optional[Any]:
         if self.enabled:
             fn(self, *args, **kwargs)
 
@@ -36,7 +41,7 @@ def wrapped_fn(self, *args, **kwargs):
 
 class InternalDebugger(object):
 
-    def __init__(self, trainer):
+    def __init__(self, trainer: 'pl.Trainer') -> None:
         self.enabled = os.environ.get('PL_DEV_DEBUG', '0') == '1'
         self.trainer = trainer
         self.logged_metrics = []
@@ -56,7 +61,7 @@ def __init__(self, trainer):
     def track_event(
         self,
         evt_type: str,
-        evt_value: Any = None,
+        evt_value: Optional[Any] = None,
         global_rank: Optional[int] = None,
         local_rank: Optional[int] = None,
         comment: str = ''
@@ -70,7 +75,7 @@ def track_event(
             "comment": comment,
         })
 
-    def count_events(self, evt_type: str, strict=False) -> int:
+    def count_events(self, evt_type: str, strict: bool = False) -> int:
         count = 0
         for evt in self.events:
             if strict and evt["event"] == evt_type:
@@ -80,7 +85,7 @@ def count_events(self, evt_type: str, strict=False) -> int:
         return count
 
     @enabled_only
-    def track_load_dataloader_call(self, name, dataloaders):
+    def track_load_dataloader_call(self, name: str, dataloaders: List[DataLoader]):
         loader_counts = len(dataloaders)
 
         lengths = []
@@ -111,18 +116,25 @@ def track_load_dataloader_call(self, name, dataloaders):
             self.test_dataloader_calls.append(values)
 
     @enabled_only
-    def track_logged_metrics_history(self, scalar_metrics):
+    def track_logged_metrics_history(self, scalar_metrics: Dict[str, torch.Tensor]) -> None:
         scalar_metrics['global_step'] = self.trainer.global_step
         self.logged_metrics.append(scalar_metrics)
 
     @enabled_only
-    def track_train_loss_history(self, batch_idx, loss):
+    def track_train_loss_history(self, batch_idx: int, loss: torch.Tensor) -> None:
         loss_dict = {'batch_idx': batch_idx, 'epoch': self.trainer.current_epoch, 'loss': loss.detach()}
         self.saved_train_losses.append(loss_dict)
 
     @enabled_only
     def track_lr_schedulers_update(
-        self, batch_idx, interval, scheduler_idx, old_lr, new_lr, monitor_key=None, monitor_val=None
+        self,
+        batch_idx: int,
+        interval: int,
+        scheduler_idx: int,
+        old_lr: float,
+        new_lr: float,
+        monitor_key: Optional[str] = None,
+        monitor_val: Optional[torch.Tensor] = None
     ):
         loss_dict = {
             'batch_idx': batch_idx,
@@ -137,7 +149,7 @@ def track_lr_schedulers_update(
         self.saved_lr_scheduler_updates.append(loss_dict)
 
     @enabled_only
-    def track_eval_loss_history(self, batch_idx, dataloader_idx, output):
+    def track_eval_loss_history(self, batch_idx: int, dataloader_idx: int, output: torch.Tensor):
         loss_dict = {
             'sanity_check': self.trainer.sanity_checking,
             'dataloader_idx': dataloader_idx,
@@ -152,12 +164,12 @@ def track_eval_loss_history(self, batch_idx, dataloader_idx, output):
             self.saved_val_losses.append(loss_dict)
 
     @enabled_only
-    def track_pbar_metrics_history(self, metrics):
+    def track_pbar_metrics_history(self, metrics: Dict[str, torch.Tensor]) -> None:
         metrics['debug_epoch'] = self.trainer.current_epoch
         self.pbar_added_metrics.append(metrics)
 
     @enabled_only
-    def track_early_stopping_history(self, callback, current):
+    def track_early_stopping_history(self, callback: 'pl.Callback', current: torch.Tensor) -> None:
         debug_dict = {
             'epoch': self.trainer.current_epoch,
             'global_step': self.trainer.global_step,
@@ -169,7 +181,7 @@ def track_early_stopping_history(self, callback, current):
         self.early_stopping_history.append(debug_dict)
 
     @enabled_only
-    def track_checkpointing_history(self, filepath):
+    def track_checkpointing_history(self, filepath: str) -> None:
         cb = self.trainer.checkpoint_callback
         debug_dict = {
             'epoch': self.trainer.current_epoch,
@@ -181,12 +193,12 @@ def track_checkpointing_history(self, filepath):
         self.checkpoint_callback_history.append(debug_dict)
 
     @property
-    def num_seen_sanity_check_batches(self):
+    def num_seen_sanity_check_batches(self) -> int:
         count = len([x for x in self.saved_val_losses if x['sanity_check']])
         return count
 
     @property
-    def num_seen_val_check_batches(self):
+    def num_seen_val_check_batches(self) -> Counter:
         counts = Counter()
         for x in self.saved_val_losses:
             if not x['sanity_check']:
@@ -194,7 +206,7 @@ def num_seen_val_check_batches(self):
         return counts
 
     @property
-    def num_seen_test_check_batches(self):
+    def num_seen_test_check_batches(self) -> Counter:
         counts = Counter()
         for x in self.saved_test_losses:
             if not x['sanity_check']:

From 31a8315bad4deadadb361e1723b2e178ac421520 Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Mon, 26 Apr 2021 21:01:34 +0200
Subject: [PATCH 06/51] Add typing for device_dtype_mixin.py

---
 pytorch_lightning/utilities/device_dtype_mixin.py | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/pytorch_lightning/utilities/device_dtype_mixin.py b/pytorch_lightning/utilities/device_dtype_mixin.py
index f08a421ce1844..77c48d36d5cfd 100644
--- a/pytorch_lightning/utilities/device_dtype_mixin.py
+++ b/pytorch_lightning/utilities/device_dtype_mixin.py
@@ -12,7 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-from typing import Optional, Union
+from typing import Any, Optional, Union
 import torch
 from torch.nn import Module
 
@@ -22,7 +22,7 @@
 class DeviceDtypeModuleMixin(Module):
     __jit_unused_properties__ = ['device', 'dtype']
 
-    def __init__(self):
+    def __init__(self) -> None:
         super().__init__()
         self._dtype = torch.get_default_dtype()
         self._device = torch.device('cpu')
@@ -47,7 +47,7 @@ def device(self) -> Union[str, torch.device]:
         return device
 
     @parameter_validation
-    def to(self, *args, **kwargs) -> Module:
+    def to(self, *args: Any, **kwargs: Any) -> Module:
         """Moves and/or casts the parameters and buffers.
 
         This can be called as
@@ -178,7 +178,7 @@ def half(self) -> Module:
         self.__update_properties(dtype=torch.half)
         return super().half()
 
-    def __update_properties(self, device: Optional[torch.device] = None, dtype: Optional[torch.dtype] = None):
+    def __update_properties(self, device: Optional[torch.device] = None, dtype: Optional[torch.dtype] = None) -> None:
 
         def apply_fn(module):
             if not isinstance(module, DeviceDtypeModuleMixin):

From cf1240cf8f3040341455f97f52002ccc13d9d350 Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Mon, 26 Apr 2021 21:04:39 +0200
Subject: [PATCH 07/51] Add typing for device_parser

---
 pytorch_lightning/utilities/device_parser.py | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/pytorch_lightning/utilities/device_parser.py b/pytorch_lightning/utilities/device_parser.py
index f81a4ece1c6d0..dee559b717f60 100644
--- a/pytorch_lightning/utilities/device_parser.py
+++ b/pytorch_lightning/utilities/device_parser.py
@@ -11,7 +11,7 @@
 # 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 typing import Any, List, MutableSequence, Optional, Tuple, Union
+from typing import Any, List, MutableSequence, Optional, Set, Tuple, Union
 
 import torch
 
@@ -171,13 +171,13 @@ def _check_data_type(device_ids: Any) -> None:
         raise MisconfigurationException("Device ID's (GPU/TPU) must be int, string or sequence of ints or None.")
 
 
-def _tpu_cores_valid(tpu_cores):
+def _tpu_cores_valid(tpu_cores: Optional[Union[int, List[int], Tuple[int], Set[int]]]) -> bool:
     # allow 1 or 8 cores
     if tpu_cores in (1, 8, None):
         return True
 
     # allow picking 1 of 8 indexes
-    if isinstance(tpu_cores, (list, tuple, set)):
+    if isinstance(tpu_cores, (List, Tuple, Set)):
         has_1_tpu_idx = len(tpu_cores) == 1
         is_valid_tpu_idx = tpu_cores[0] in range(1, 9)
 
@@ -187,7 +187,7 @@ def _tpu_cores_valid(tpu_cores):
     return False
 
 
-def _parse_tpu_cores_str(tpu_cores):
+def _parse_tpu_cores_str(tpu_cores: str) -> Union[int, List[int]]:
     if tpu_cores in ('1', '8'):
         tpu_cores = int(tpu_cores)
     else:

From 65d3bd5d9d46d639abcb9d1969da02501aa02d4e Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Mon, 26 Apr 2021 21:14:12 +0200
Subject: [PATCH 08/51] Add typing for distributed

---
 pytorch_lightning/utilities/distributed.py | 28 +++++++++++-----------
 1 file changed, 14 insertions(+), 14 deletions(-)

diff --git a/pytorch_lightning/utilities/distributed.py b/pytorch_lightning/utilities/distributed.py
index a54d00a983d9e..1df011f80e38b 100644
--- a/pytorch_lightning/utilities/distributed.py
+++ b/pytorch_lightning/utilities/distributed.py
@@ -16,7 +16,7 @@
 import os
 import warnings
 from functools import partial, wraps
-from typing import Any, Optional, Union
+from typing import Any, Callable, List, Optional, Sequence, Tuple, Union
 
 import torch
 from torch.nn.parallel.distributed import DistributedDataParallel
@@ -41,10 +41,10 @@ class group:
 log = logging.getLogger(__name__)
 
 
-def rank_zero_only(fn):
+def rank_zero_only(fn: Callable) -> Optional[Callable]:
 
     @wraps(fn)
-    def wrapped_fn(*args, **kwargs):
+    def wrapped_fn(*args: Any, **kwargs: Any) -> Optional[Any]:
         if rank_zero_only.rank == 0:
             return fn(*args, **kwargs)
 
@@ -65,15 +65,15 @@ def _get_rank() -> int:
 rank_zero_only.rank = getattr(rank_zero_only, 'rank', _get_rank())
 
 
-def _warn(*args, **kwargs):
+def _warn(*args: Any, **kwargs: Any) -> None:
     warnings.warn(*args, **kwargs)
 
 
-def _info(*args, **kwargs):
+def _info(*args: Any, **kwargs: Any) -> None:
     log.info(*args, **kwargs)
 
 
-def _debug(*args, **kwargs):
+def _debug(*args: Any, **kwargs: Any) -> None:
     log.debug(*args, **kwargs)
 
 
@@ -83,7 +83,7 @@ def _debug(*args, **kwargs):
 rank_zero_deprecation = partial(rank_zero_warn, category=DeprecationWarning)
 
 
-def gather_all_tensors(result: Union[torch.Tensor], group: Optional[Any] = None):
+def gather_all_tensors(result: Union[torch.Tensor], group: Optional[Any] = None) -> List[torch.Tensor]:
     """
     Function to gather all tensors from several ddp processes onto a list that
     is broadcasted to all processes
@@ -114,7 +114,7 @@ def gather_all_tensors(result: Union[torch.Tensor], group: Optional[Any] = None)
 
 
 def sync_ddp_if_available(
-    result: Union[torch.Tensor],
+    result: torch.Tensor,
     group: Optional[Any] = None,
     reduce_op: Optional[Union[ReduceOp, str]] = None
 ) -> torch.Tensor:
@@ -135,7 +135,7 @@ def sync_ddp_if_available(
 
 
 def sync_ddp(
-    result: Union[torch.Tensor],
+    result: torch.Tensor,
     group: Optional[Any] = None,
     reduce_op: Optional[Union[ReduceOp, str]] = None
 ) -> torch.Tensor:
@@ -174,7 +174,7 @@ def sync_ddp(
 class AllGatherGrad(torch.autograd.Function):
 
     @staticmethod
-    def forward(ctx, tensor, group=group.WORLD):
+    def forward(ctx: object, tensor: torch.Tensor, group=group.WORLD) -> torch.Tensor:
         ctx.group = group
 
         gathered_tensor = [torch.zeros_like(tensor) for _ in range(torch.distributed.get_world_size())]
@@ -185,7 +185,7 @@ def forward(ctx, tensor, group=group.WORLD):
         return gathered_tensor
 
     @staticmethod
-    def backward(ctx, *grad_output):
+    def backward(ctx: object, *grad_output: Sequence[torch.Tensor]) -> Tuple[torch.Tensor, None]:
         grad_output = torch.cat(grad_output)
 
         torch.distributed.all_reduce(grad_output, op=torch.distributed.ReduceOp.SUM, async_op=False, group=ctx.group)
@@ -194,7 +194,7 @@ def backward(ctx, *grad_output):
 
 
 def all_gather_ddp_if_available(
-    tensor: Union[torch.Tensor], group: Optional[Any] = None, sync_grads: bool = False
+    tensor: torch.Tensor, group: Optional[Any] = None, sync_grads: bool = False
 ) -> torch.Tensor:
     """
     Function to gather a tensor from several distributed processes
@@ -220,8 +220,8 @@ def all_gather_ddp_if_available(
 def register_ddp_comm_hook(
     model: DistributedDataParallel,
     ddp_comm_state: Optional[object] = None,
-    ddp_comm_hook: Optional[callable] = None,
-    ddp_comm_wrapper: Optional[callable] = None,
+    ddp_comm_hook: Optional[Callable] = None,
+    ddp_comm_wrapper: Optional[Callable] = None,
 ) -> None:
     """
     Function to register communication hook for DDP model

From e67d4c771a19fb8b71262b7e069b00c1172595a0 Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Mon, 26 Apr 2021 21:18:29 +0200
Subject: [PATCH 09/51] Add typing for imports

---
 pytorch_lightning/utilities/imports.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/pytorch_lightning/utilities/imports.py b/pytorch_lightning/utilities/imports.py
index 791cef7ff2665..4b92a277e0ca7 100644
--- a/pytorch_lightning/utilities/imports.py
+++ b/pytorch_lightning/utilities/imports.py
@@ -17,6 +17,7 @@
 import platform
 import sys
 from importlib.util import find_spec
+from typing import Callable
 
 import torch
 from packaging.version import Version
@@ -42,7 +43,7 @@ def _module_available(module_path: str) -> bool:
         return False
 
 
-def _compare_version(package: str, op, version) -> bool:
+def _compare_version(package: str, op: Callable, version: str) -> bool:
     """
     Compare package version with some requirements
 

From a854722e090f2e5f828421cc009fab94a3da1282 Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Mon, 26 Apr 2021 21:24:11 +0200
Subject: [PATCH 10/51] Add typing for memory

---
 pytorch_lightning/utilities/memory.py | 16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)

diff --git a/pytorch_lightning/utilities/memory.py b/pytorch_lightning/utilities/memory.py
index 6c01390a8c81e..0ee5e4ac4c9ce 100644
--- a/pytorch_lightning/utilities/memory.py
+++ b/pytorch_lightning/utilities/memory.py
@@ -13,11 +13,15 @@
 # limitations under the License.
 
 import gc
+from typing import Any, Dict, Union
 
 import torch
 
 
-def recursive_detach(in_dict: dict, to_cpu: bool = False) -> dict:
+def recursive_detach(
+    in_dict: Dict[Any, Union[torch.Tensor, Dict[Any, torch.Tensor]]],
+    to_cpu: bool = False
+) -> Dict[Any, torch.Tensor]:
     """Detach all tensors in `in_dict`.
 
     May operate recursively if some of the values in `in_dict` are dictionaries
@@ -43,14 +47,14 @@ def recursive_detach(in_dict: dict, to_cpu: bool = False) -> dict:
     return out_dict
 
 
-def is_oom_error(exception):
+def is_oom_error(exception: Any) -> bool:
     return is_cuda_out_of_memory(exception) \
         or is_cudnn_snafu(exception) \
         or is_out_of_cpu_memory(exception)
 
 
 # based on https://github.com/BlackHC/toma/blob/master/toma/torch_cuda_memory.py
-def is_cuda_out_of_memory(exception):
+def is_cuda_out_of_memory(exception: Any) -> bool:
     return isinstance(exception, RuntimeError) \
         and len(exception.args) == 1 \
         and "CUDA" in exception.args[0] \
@@ -58,7 +62,7 @@ def is_cuda_out_of_memory(exception):
 
 
 # based on https://github.com/BlackHC/toma/blob/master/toma/torch_cuda_memory.py
-def is_cudnn_snafu(exception):
+def is_cudnn_snafu(exception: Any) -> bool:
     # For/because of https://github.com/pytorch/pytorch/issues/4107
     return isinstance(exception, RuntimeError) \
         and len(exception.args) == 1 \
@@ -66,14 +70,14 @@ def is_cudnn_snafu(exception):
 
 
 # based on https://github.com/BlackHC/toma/blob/master/toma/cpu_memory.py
-def is_out_of_cpu_memory(exception):
+def is_out_of_cpu_memory(exception: Any) -> bool:
     return isinstance(exception, RuntimeError) \
         and len(exception.args) == 1 \
         and "DefaultCPUAllocator: can't allocate memory" in exception.args[0]
 
 
 # based on https://github.com/BlackHC/toma/blob/master/toma/torch_cuda_memory.py
-def garbage_collection_cuda():
+def garbage_collection_cuda() -> None:
     """Garbage collection Torch (CUDA) memory."""
     gc.collect()
     if torch.cuda.is_available():

From ff7c24a7f94ed836ef9f52829e332d8313a770be Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Mon, 26 Apr 2021 21:28:40 +0200
Subject: [PATCH 11/51] Add typing for metrics

---
 pytorch_lightning/utilities/metrics.py | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/pytorch_lightning/utilities/metrics.py b/pytorch_lightning/utilities/metrics.py
index bd57470dc270e..e547e69894a05 100644
--- a/pytorch_lightning/utilities/metrics.py
+++ b/pytorch_lightning/utilities/metrics.py
@@ -13,12 +13,14 @@
 # limitations under the License.
 """Helper functions to operate on metric values. """
 
+from typing import Dict, Union
+
 import torch
 
 from pytorch_lightning.utilities.exceptions import MisconfigurationException
 
 
-def metrics_to_scalars(metrics: dict) -> dict:
+def metrics_to_scalars(metrics: Dict[str, Union[torch.Tensor, dict]]) -> Dict[str, float]:
     """ Recursively walk through a dictionary of metrics and convert single-item tensors to scalar values. """
 
     # TODO: this is duplicated in MetricsHolder. should be unified

From bbb01c4a7f4987ae7faaf76b0a544c0ea6adf020 Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Mon, 26 Apr 2021 21:30:36 +0200
Subject: [PATCH 12/51] Add typing for model_helpers

---
 pytorch_lightning/utilities/model_helpers.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/pytorch_lightning/utilities/model_helpers.py b/pytorch_lightning/utilities/model_helpers.py
index 87bd9e6c4545d..6aaeb96dec599 100644
--- a/pytorch_lightning/utilities/model_helpers.py
+++ b/pytorch_lightning/utilities/model_helpers.py
@@ -14,11 +14,12 @@
 
 from typing import Union
 
+import pytorch_lightning as pl
 from pytorch_lightning.core.datamodule import LightningDataModule
 from pytorch_lightning.core.lightning import LightningModule
 
 
-def is_overridden(method_name: str, model: Union[LightningModule, LightningDataModule]) -> bool:
+def is_overridden(method_name: str, model: Union['pl.LightningModule', 'pl.LightningDataModule']) -> bool:
     # if you pass DataModule instead of None or a LightningModule, we use LightningDataModule as super
     # TODO - refector this function to accept model_name, instance, parent so it makes more sense
     super_object = LightningModule if not isinstance(model, LightningDataModule) else LightningDataModule

From b2cdfc1987c84723d16e3b6c434d548a29d9c826 Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Mon, 26 Apr 2021 21:45:55 +0200
Subject: [PATCH 13/51] Add typing for parsing

---
 pytorch_lightning/utilities/parsing.py | 31 +++++++++++++-------------
 1 file changed, 16 insertions(+), 15 deletions(-)

diff --git a/pytorch_lightning/utilities/parsing.py b/pytorch_lightning/utilities/parsing.py
index ae83ba15a9c52..9fdc5f20432ab 100644
--- a/pytorch_lightning/utilities/parsing.py
+++ b/pytorch_lightning/utilities/parsing.py
@@ -16,8 +16,9 @@
 import pickle
 import types
 from argparse import Namespace
-from typing import Any, Dict, Optional, Sequence, Tuple, Union
+from typing import Any, Dict, List, Optional, Sequence, Tuple, Type, Union
 
+import pytorch_lightning as pl
 from pytorch_lightning.utilities import rank_zero_warn
 
 
@@ -66,7 +67,7 @@ def is_picklable(obj: object) -> bool:
         return False
 
 
-def clean_namespace(hparams):
+def clean_namespace(hparams: Union[Dict[str, Any], Namespace]) -> None:
     """Removes all unpicklable entries from hparams"""
 
     hparams_dict = hparams
@@ -96,7 +97,7 @@ def parse_class_init_keys(cls) -> Tuple[str, str, str]:
     # self is always first
     n_self = init_params[0].name
 
-    def _get_first_if_any(params, param_type):
+    def _get_first_if_any(params: Sequence[Namespace], param_type: Type) -> str:
         for p in params:
             if p.kind == param_type:
                 return p.name
@@ -107,7 +108,7 @@ def _get_first_if_any(params, param_type):
     return n_self, n_args, n_kwargs
 
 
-def get_init_args(frame) -> dict:
+def get_init_args(frame) -> Dict[str, Any]:
     _, _, _, local_vars = inspect.getargvalues(frame)
     if '__class__' not in local_vars:
         return {}
@@ -123,7 +124,7 @@ def get_init_args(frame) -> dict:
     return local_args
 
 
-def collect_init_args(frame, path_args: list, inside: bool = False) -> list:
+def collect_init_args(frame, path_args: List[Dict[str, Any]], inside: bool = False) -> List[Dict[str, Any]]:
     """
     Recursively collects the arguments passed to the child constructors in the inheritance tree.
 
@@ -149,7 +150,7 @@ def collect_init_args(frame, path_args: list, inside: bool = False) -> list:
         return path_args
 
 
-def flatten_dict(source, result=None):
+def flatten_dict(source: Dict[str, Any], result=None) -> Dict[str, Any]:
     if result is None:
         result = {}
 
@@ -164,7 +165,7 @@ def flatten_dict(source, result=None):
 
 def save_hyperparameters(
     obj: Any,
-    *args,
+    *args: Any,
     ignore: Optional[Union[Sequence[str], str]] = None,
     frame: Optional[types.FrameType] = None
 ) -> None:
@@ -219,16 +220,16 @@ class AttributeDict(Dict):
     "my-key":  3.14
     """
 
-    def __getattr__(self, key):
+    def __getattr__(self, key: str) -> Optional[Any]:
         try:
             return self[key]
         except KeyError as exp:
             raise AttributeError(f'Missing attribute "{key}"') from exp
 
-    def __setattr__(self, key, val):
+    def __setattr__(self, key: str, val: Any) -> None:
         self[key] = val
 
-    def __repr__(self):
+    def __repr__(self) -> str:
         if not len(self):
             return ""
         max_key_length = max([len(str(k)) for k in self])
@@ -238,7 +239,7 @@ def __repr__(self):
         return out
 
 
-def _lightning_get_all_attr_holders(model, attribute):
+def _lightning_get_all_attr_holders(model: 'pl.LightningModule', attribute: str) -> Any:
     """
     Special attribute finding for Lightning. Gets all of the objects or dicts that holds attribute.
     Checks for attribute in model namespace, the old hparams namespace/dict, and the datamodule.
@@ -263,7 +264,7 @@ def _lightning_get_all_attr_holders(model, attribute):
     return holders
 
 
-def _lightning_get_first_attr_holder(model, attribute):
+def _lightning_get_first_attr_holder(model: 'pl.LightningModule', attribute: str) -> Optional[Any]:
     """
     Special attribute finding for Lightning.  Gets the object or dict that holds attribute, or None.
     Checks for attribute in model namespace, the old hparams namespace/dict, and the datamodule,
@@ -276,7 +277,7 @@ def _lightning_get_first_attr_holder(model, attribute):
     return holders[-1]
 
 
-def lightning_hasattr(model, attribute):
+def lightning_hasattr(model: 'pl.LightningModule', attribute: str) -> bool:
     """
     Special hasattr for Lightning. Checks for attribute in model namespace,
     the old hparams namespace/dict, and the datamodule.
@@ -284,7 +285,7 @@ def lightning_hasattr(model, attribute):
     return _lightning_get_first_attr_holder(model, attribute) is not None
 
 
-def lightning_getattr(model, attribute):
+def lightning_getattr(model: 'pl.LightningModule', attribute: str) -> Any:
     """
     Special getattr for Lightning. Checks for attribute in model namespace,
     the old hparams namespace/dict, and the datamodule.
@@ -306,7 +307,7 @@ def lightning_getattr(model, attribute):
     return getattr(holder, attribute)
 
 
-def lightning_setattr(model, attribute, value):
+def lightning_setattr(model: 'pl.LightningModule', attribute: str, value: Any) -> None:
     """
     Special setattr for Lightning. Checks for attribute in model namespace
     and the old hparams namespace/dict.

From 900d9949af7561eac5ad51a8927e97edd6f76936 Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Mon, 26 Apr 2021 21:46:08 +0200
Subject: [PATCH 14/51] Add typing for parsing

---
 pytorch_lightning/utilities/parsing.py | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/pytorch_lightning/utilities/parsing.py b/pytorch_lightning/utilities/parsing.py
index 9fdc5f20432ab..12d9f68597ae7 100644
--- a/pytorch_lightning/utilities/parsing.py
+++ b/pytorch_lightning/utilities/parsing.py
@@ -108,7 +108,7 @@ def _get_first_if_any(params: Sequence[Namespace], param_type: Type) -> str:
     return n_self, n_args, n_kwargs
 
 
-def get_init_args(frame) -> Dict[str, Any]:
+def get_init_args(frame: object) -> Dict[str, Any]:
     _, _, _, local_vars = inspect.getargvalues(frame)
     if '__class__' not in local_vars:
         return {}
@@ -124,7 +124,7 @@ def get_init_args(frame) -> Dict[str, Any]:
     return local_args
 
 
-def collect_init_args(frame, path_args: List[Dict[str, Any]], inside: bool = False) -> List[Dict[str, Any]]:
+def collect_init_args(frame:object, path_args: List[Dict[str, Any]], inside: bool = False) -> List[Dict[str, Any]]:
     """
     Recursively collects the arguments passed to the child constructors in the inheritance tree.
 
@@ -150,7 +150,7 @@ def collect_init_args(frame, path_args: List[Dict[str, Any]], inside: bool = Fal
         return path_args
 
 
-def flatten_dict(source: Dict[str, Any], result=None) -> Dict[str, Any]:
+def flatten_dict(source: Dict[str, Any], result: Optioanl[Dict[str, Any]] = None) -> Dict[str, Any]:
     if result is None:
         result = {}
 

From 24b0125ff1c524813eddf7fd24ad111443fcc66a Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Mon, 26 Apr 2021 21:56:52 +0200
Subject: [PATCH 15/51] Add typing to multiple files and fix a typo

* Files: seed, upgrade_checkpoint, warnings

* Fix typo in parsing
---
 pytorch_lightning/utilities/parsing.py            | 2 +-
 pytorch_lightning/utilities/seed.py               | 2 +-
 pytorch_lightning/utilities/upgrade_checkpoint.py | 2 +-
 pytorch_lightning/utilities/warnings.py           | 8 +++++---
 4 files changed, 8 insertions(+), 6 deletions(-)

diff --git a/pytorch_lightning/utilities/parsing.py b/pytorch_lightning/utilities/parsing.py
index 12d9f68597ae7..4224e1c7f6630 100644
--- a/pytorch_lightning/utilities/parsing.py
+++ b/pytorch_lightning/utilities/parsing.py
@@ -150,7 +150,7 @@ def collect_init_args(frame:object, path_args: List[Dict[str, Any]], inside: boo
         return path_args
 
 
-def flatten_dict(source: Dict[str, Any], result: Optioanl[Dict[str, Any]] = None) -> Dict[str, Any]:
+def flatten_dict(source: Dict[str, Any], result: Optional[Dict[str, Any]] = None) -> Dict[str, Any]:
     if result is None:
         result = {}
 
diff --git a/pytorch_lightning/utilities/seed.py b/pytorch_lightning/utilities/seed.py
index b7eaba72c1b02..0fc149d0b7d04 100644
--- a/pytorch_lightning/utilities/seed.py
+++ b/pytorch_lightning/utilities/seed.py
@@ -78,7 +78,7 @@ def _select_seed_randomly(min_seed_value: int = 0, max_seed_value: int = 255) ->
     return random.randint(min_seed_value, max_seed_value)
 
 
-def pl_worker_init_function(worker_id: int, rank: Optional = None) -> None:  # pragma: no cover
+def pl_worker_init_function(worker_id: int, rank: Optional[int] = None) -> None:  # pragma: no cover
     """
     The worker_init_fn that Lightning automatically adds to your dataloader if you previously set
     set the seed with ``seed_everything(seed, workers=True)``.
diff --git a/pytorch_lightning/utilities/upgrade_checkpoint.py b/pytorch_lightning/utilities/upgrade_checkpoint.py
index 4896845f10263..d83a682511188 100644
--- a/pytorch_lightning/utilities/upgrade_checkpoint.py
+++ b/pytorch_lightning/utilities/upgrade_checkpoint.py
@@ -30,7 +30,7 @@
 log = logging.getLogger(__name__)
 
 
-def upgrade_checkpoint(filepath):
+def upgrade_checkpoint(filepath: str) -> None:
     checkpoint = torch.load(filepath)
     checkpoint["callbacks"] = checkpoint.get("callbacks") or {}
 
diff --git a/pytorch_lightning/utilities/warnings.py b/pytorch_lightning/utilities/warnings.py
index a3dde95fa928f..6dfb67133cf19 100644
--- a/pytorch_lightning/utilities/warnings.py
+++ b/pytorch_lightning/utilities/warnings.py
@@ -11,18 +11,20 @@
 # 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 typing import Any
+
 from pytorch_lightning.utilities.distributed import rank_zero_warn
 
 
 class WarningCache:
 
-    def __init__(self):
+    def __init__(self) -> None:
         self.warnings = set()
 
-    def warn(self, m, *args, **kwargs):
+    def warn(self, m: Any, *args: Any, **kwargs: Any) -> None:
         if m not in self.warnings:
             self.warnings.add(m)
             rank_zero_warn(m, *args, **kwargs)
 
-    def clear(self):
+    def clear(self) -> None:
         self.warnings.clear()

From 7ca60247c253540e67747595ec3a13d0dd4be6d9 Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Mon, 26 Apr 2021 22:03:45 +0200
Subject: [PATCH 16/51] Add typing for xla_device

---
 pytorch_lightning/utilities/xla_device.py | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/pytorch_lightning/utilities/xla_device.py b/pytorch_lightning/utilities/xla_device.py
index 49ec176d4cdbb..8e2836c59bdae 100644
--- a/pytorch_lightning/utilities/xla_device.py
+++ b/pytorch_lightning/utilities/xla_device.py
@@ -16,6 +16,7 @@
 import queue as q
 import traceback
 from multiprocessing import Process, Queue
+from typing import Any, Callable, Union
 
 from pytorch_lightning.utilities.imports import _XLA_AVAILABLE
 
@@ -26,7 +27,7 @@
 TPU_CHECK_TIMEOUT = 25
 
 
-def inner_f(queue, func, *args, **kwargs):  # pragma: no cover
+def inner_f(queue: Queue, func: Callable, *args: Any, **kwargs: Any) -> None:  # pragma: no cover
     try:
         queue.put(func(*args, **kwargs))
     # todo: specify the possible exception
@@ -35,10 +36,10 @@ def inner_f(queue, func, *args, **kwargs):  # pragma: no cover
         queue.put(None)
 
 
-def pl_multi_process(func):
+def pl_multi_process(func: Callable) -> Callable:
 
     @functools.wraps(func)
-    def wrapper(*args, **kwargs):
+    def wrapper(*args: Any, **kwargs: Any) -> Union[Any, bool]:
         queue = Queue()
         proc = Process(target=inner_f, args=(queue, func, *args), kwargs=kwargs)
         proc.start()

From 7576c450146d98a83c78d30a4c0f430088e82331 Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Mon, 26 Apr 2021 22:25:08 +0200
Subject: [PATCH 17/51] Add missing whitespace afer ':' in parsing.py

---
 pytorch_lightning/utilities/parsing.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pytorch_lightning/utilities/parsing.py b/pytorch_lightning/utilities/parsing.py
index 4224e1c7f6630..765b548932675 100644
--- a/pytorch_lightning/utilities/parsing.py
+++ b/pytorch_lightning/utilities/parsing.py
@@ -124,7 +124,7 @@ def get_init_args(frame: object) -> Dict[str, Any]:
     return local_args
 
 
-def collect_init_args(frame:object, path_args: List[Dict[str, Any]], inside: bool = False) -> List[Dict[str, Any]]:
+def collect_init_args(frame: object, path_args: List[Dict[str, Any]], inside: bool = False) -> List[Dict[str, Any]]:
     """
     Recursively collects the arguments passed to the child constructors in the inheritance tree.
 

From 2cee78f75af4d941ff53fb1c17a6fe813637cd18 Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Wed, 28 Apr 2021 16:37:53 +0200
Subject: [PATCH 18/51] Add some missing typing

---
 pytorch_lightning/utilities/cli.py                | 4 ++--
 pytorch_lightning/utilities/debugging.py          | 6 +++---
 pytorch_lightning/utilities/device_dtype_mixin.py | 2 +-
 pytorch_lightning/utilities/distributed.py        | 2 +-
 pytorch_lightning/utilities/finite_checks.py      | 6 +++---
 pytorch_lightning/utilities/memory.py             | 2 +-
 pytorch_lightning/utilities/seed.py               | 2 +-
 7 files changed, 12 insertions(+), 12 deletions(-)

diff --git a/pytorch_lightning/utilities/cli.py b/pytorch_lightning/utilities/cli.py
index 730fe2798ef24..7b306d58dc4ed 100644
--- a/pytorch_lightning/utilities/cli.py
+++ b/pytorch_lightning/utilities/cli.py
@@ -97,8 +97,8 @@ def __init__(
         datamodule_class: Optional['pl.LightningDataModule'] = None,
         save_config_callback: SaveConfigCallback = SaveConfigCallback,
         trainer_class: 'pl.Trainer' = Trainer,
-        trainer_defaults: Dict[str, Any] = None,
-        seed_everything_default: int = None,
+        trainer_defaults: Optional[Dict[str, Any]] = None,
+        seed_everything_default: Optional[int] = None,
         description: str = 'pytorch-lightning trainer command line tool',
         env_prefix: str = 'PL',
         env_parse: bool = False,
diff --git a/pytorch_lightning/utilities/debugging.py b/pytorch_lightning/utilities/debugging.py
index 58fb1aeffc2ef..d441b025388fe 100644
--- a/pytorch_lightning/utilities/debugging.py
+++ b/pytorch_lightning/utilities/debugging.py
@@ -85,7 +85,7 @@ def count_events(self, evt_type: str, strict: bool = False) -> int:
         return count
 
     @enabled_only
-    def track_load_dataloader_call(self, name: str, dataloaders: List[DataLoader]):
+    def track_load_dataloader_call(self, name: str, dataloaders: List[DataLoader]) -> None:
         loader_counts = len(dataloaders)
 
         lengths = []
@@ -135,7 +135,7 @@ def track_lr_schedulers_update(
         new_lr: float,
         monitor_key: Optional[str] = None,
         monitor_val: Optional[torch.Tensor] = None
-    ):
+    ) -> None:
         loss_dict = {
             'batch_idx': batch_idx,
             'interval': interval,
@@ -149,7 +149,7 @@ def track_lr_schedulers_update(
         self.saved_lr_scheduler_updates.append(loss_dict)
 
     @enabled_only
-    def track_eval_loss_history(self, batch_idx: int, dataloader_idx: int, output: torch.Tensor):
+    def track_eval_loss_history(self, batch_idx: int, dataloader_idx: int, output: torch.Tensor) -> None:
         loss_dict = {
             'sanity_check': self.trainer.sanity_checking,
             'dataloader_idx': dataloader_idx,
diff --git a/pytorch_lightning/utilities/device_dtype_mixin.py b/pytorch_lightning/utilities/device_dtype_mixin.py
index 77c48d36d5cfd..27d2b352bf4ff 100644
--- a/pytorch_lightning/utilities/device_dtype_mixin.py
+++ b/pytorch_lightning/utilities/device_dtype_mixin.py
@@ -32,7 +32,7 @@ def dtype(self) -> Union[str, torch.dtype]:
         return self._dtype
 
     @dtype.setter
-    def dtype(self, new_dtype: Union[str, torch.dtype]):
+    def dtype(self, new_dtype: Union[str, torch.dtype]) -> None:
         # necessary to avoid infinite recursion
         raise RuntimeError('Cannot set the dtype explicitly. Please use module.to(new_dtype).')
 
diff --git a/pytorch_lightning/utilities/distributed.py b/pytorch_lightning/utilities/distributed.py
index 1df011f80e38b..7cee9efa2c052 100644
--- a/pytorch_lightning/utilities/distributed.py
+++ b/pytorch_lightning/utilities/distributed.py
@@ -83,7 +83,7 @@ def _debug(*args: Any, **kwargs: Any) -> None:
 rank_zero_deprecation = partial(rank_zero_warn, category=DeprecationWarning)
 
 
-def gather_all_tensors(result: Union[torch.Tensor], group: Optional[Any] = None) -> List[torch.Tensor]:
+def gather_all_tensors(result: torch.Tensor, group: Optional[Any] = None) -> List[torch.Tensor]:
     """
     Function to gather all tensors from several ddp processes onto a list that
     is broadcasted to all processes
diff --git a/pytorch_lightning/utilities/finite_checks.py b/pytorch_lightning/utilities/finite_checks.py
index 770ea7a2276f0..c67958ad74f67 100644
--- a/pytorch_lightning/utilities/finite_checks.py
+++ b/pytorch_lightning/utilities/finite_checks.py
@@ -16,19 +16,19 @@
 import logging
 
 import torch
-import torch.nn as nn
+from torch.nn import Module
 
 log = logging.getLogger(__name__)
 
 
-def print_nan_gradients(model: nn.Module) -> None:
+def print_nan_gradients(model: Module) -> None:
     """ Iterates over model parameters and prints out parameter + gradient information if NaN. """
     for param in model.parameters():
         if (param.grad is not None) and torch.isnan(param.grad.float()).any():
             log.info(param, param.grad)
 
 
-def detect_nan_parameters(model: nn.Module) -> None:
+def detect_nan_parameters(model: Module) -> None:
     """ Iterates over model parameters and prints gradients if any parameter is not finite. """
     for name, param in model.named_parameters():
         if not torch.isfinite(param).all():
diff --git a/pytorch_lightning/utilities/memory.py b/pytorch_lightning/utilities/memory.py
index 0ee5e4ac4c9ce..5658a28bd0036 100644
--- a/pytorch_lightning/utilities/memory.py
+++ b/pytorch_lightning/utilities/memory.py
@@ -21,7 +21,7 @@
 def recursive_detach(
     in_dict: Dict[Any, Union[torch.Tensor, Dict[Any, torch.Tensor]]],
     to_cpu: bool = False
-) -> Dict[Any, torch.Tensor]:
+) -> Dict[str, torch.Tensor]:
     """Detach all tensors in `in_dict`.
 
     May operate recursively if some of the values in `in_dict` are dictionaries
diff --git a/pytorch_lightning/utilities/seed.py b/pytorch_lightning/utilities/seed.py
index 51547d5576e74..78abd06e3b289 100644
--- a/pytorch_lightning/utilities/seed.py
+++ b/pytorch_lightning/utilities/seed.py
@@ -88,7 +88,7 @@ def reset_seed() -> None:
         seed_everything(int(seed))
 
 
-def pl_worker_init_function(worker_id: int, rank: Optional = None) -> None:  # pragma: no cover
+def pl_worker_init_function(worker_id: int, rank: Optional[int] = None) -> None:  # pragma: no cover
     """
     The worker_init_fn that Lightning automatically adds to your dataloader if you previously set
     set the seed with ``seed_everything(seed, workers=True)``.

From 9b42400404f06b5f3726cb3c6a5b5c10995bedcc Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Tue, 4 May 2021 11:22:55 +0200
Subject: [PATCH 19/51] Do some fixes after reviews

* In apply_func.py device arg is marked as Optional since
torch.tensor() and other function can handle device=None

* In cloud_io.py LocalFileSystem replaced with AbstractFileSystem

* In cli.py: return Type to typing
---
 pytorch_lightning/utilities/apply_func.py | 15 ++++-----------
 pytorch_lightning/utilities/cli.py        | 10 +++++-----
 pytorch_lightning/utilities/cloud_io.py   |  4 ++--
 3 files changed, 11 insertions(+), 18 deletions(-)

diff --git a/pytorch_lightning/utilities/apply_func.py b/pytorch_lightning/utilities/apply_func.py
index d384c0edbd3bf..f501fe6284e49 100644
--- a/pytorch_lightning/utilities/apply_func.py
+++ b/pytorch_lightning/utilities/apply_func.py
@@ -36,16 +36,12 @@
 def to_dtype_tensor(
     value: Union[int, float, List[Union[int, float]]],
     dtype: Optional[torch.dtype] = None,
-    device: Union[str, torch.device] = None
+    device: Optional[Union[str, torch.device]] = None
 ) -> torch.Tensor:
-    if device is None:
-        raise MisconfigurationException("device (torch.device) should be provided.")
     return torch.tensor(value, dtype=dtype, device=device)
 
 
-def from_numpy(value: np.ndarray, device: Union[str, torch.device] = None) -> torch.Tensor:
-    if device is None:
-        raise MisconfigurationException("device (torch.device) should be provided.")
+def from_numpy(value: np.ndarray, device: Optional[Union[str, torch.device]] = None) -> torch.Tensor:
     return torch.from_numpy(value).to(device)
 
 
@@ -64,7 +60,7 @@ def apply_to_collection(
     function: Callable,
     *args: Any,
     wrong_dtype: Optional[Union[torch.dtype, Tuple[torch.dtype]]] = None,
-    **kwargs: Any
+    **kwargs: Any,
 ) -> Any:
     """
     Recursively applies a function to all elements of a certain dtype.
@@ -165,10 +161,7 @@ def batch_to(data: Any) -> Any:
     return apply_to_collection(batch, dtype=dtype, function=batch_to)
 
 
-def convert_to_tensors(data: Any, device: Union[str, torch.device] = None) -> Any:
-    if device is None:
-        raise MisconfigurationException("device (torch.device) should be provided.")
-
+def convert_to_tensors(data: Any, device: Optional[Union[str, torch.device]] = None) -> Any:
     for src_dtype, conversion_func in CONVERSION_DTYPES:
         data = apply_to_collection(data, src_dtype, partial(conversion_func, device=device))
 
diff --git a/pytorch_lightning/utilities/cli.py b/pytorch_lightning/utilities/cli.py
index 6714a36ca6122..214a41a16d407 100644
--- a/pytorch_lightning/utilities/cli.py
+++ b/pytorch_lightning/utilities/cli.py
@@ -13,7 +13,7 @@
 # limitations under the License.
 import os
 from argparse import Namespace
-from typing import Any, Dict, Optional, Union
+from typing import Any, Dict, Optional, Type, Union
 
 import pytorch_lightning as pl
 from pytorch_lightning.callbacks import Callback
@@ -93,10 +93,10 @@ class LightningCLI:
 
     def __init__(
         self,
-        model_class: 'pl.LightningModule',
-        datamodule_class: Optional['pl.LightningDataModule'] = None,
-        save_config_callback: SaveConfigCallback = SaveConfigCallback,
-        trainer_class: 'pl.Trainer' = Trainer,
+        model_class: Type['pl.LightningModule'],
+        datamodule_class: Optional[Type['pl.LightningDataModule']] = None,
+        save_config_callback: Type[SaveConfigCallback] = SaveConfigCallback,
+        trainer_class: Type['pl.Trainer'] = Trainer,
         trainer_defaults: Optional[Dict[str, Any]] = None,
         seed_everything_default: Optional[int] = None,
         description: str = 'pytorch-lightning trainer command line tool',
diff --git a/pytorch_lightning/utilities/cloud_io.py b/pytorch_lightning/utilities/cloud_io.py
index c233fbf35483f..d77c2f8745d31 100644
--- a/pytorch_lightning/utilities/cloud_io.py
+++ b/pytorch_lightning/utilities/cloud_io.py
@@ -18,7 +18,7 @@
 
 import fsspec
 import torch
-from fsspec.implementations.local import LocalFileSystem
+from fsspec.implementations.local import AbstractFileSystem, LocalFileSystem
 from packaging.version import Version
 
 
@@ -36,7 +36,7 @@ def load(
         return torch.load(f, map_location=map_location)
 
 
-def get_filesystem(path: Union[str, Path]) -> LocalFileSystem:
+def get_filesystem(path: Union[str, Path]) -> AbstractFileSystem:
     path = str(path)
     if "://" in path:
         # use the fileystem from the protocol specified

From 9cd7d19bf9473351ac798ef40c3341c58b1cda09 Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Tue, 4 May 2021 11:28:11 +0200
Subject: [PATCH 20/51] Add some missing commas'

---
 pytorch_lightning/utilities/apply_func.py  | 2 +-
 pytorch_lightning/utilities/cli.py         | 4 ++--
 pytorch_lightning/utilities/cloud_io.py    | 2 +-
 pytorch_lightning/utilities/debugging.py   | 4 ++--
 pytorch_lightning/utilities/distributed.py | 4 ++--
 pytorch_lightning/utilities/memory.py      | 2 +-
 pytorch_lightning/utilities/parsing.py     | 2 +-
 7 files changed, 10 insertions(+), 10 deletions(-)

diff --git a/pytorch_lightning/utilities/apply_func.py b/pytorch_lightning/utilities/apply_func.py
index f501fe6284e49..e000df7831b9b 100644
--- a/pytorch_lightning/utilities/apply_func.py
+++ b/pytorch_lightning/utilities/apply_func.py
@@ -36,7 +36,7 @@
 def to_dtype_tensor(
     value: Union[int, float, List[Union[int, float]]],
     dtype: Optional[torch.dtype] = None,
-    device: Optional[Union[str, torch.device]] = None
+    device: Optional[Union[str, torch.device]] = None,
 ) -> torch.Tensor:
     return torch.tensor(value, dtype=dtype, device=device)
 
diff --git a/pytorch_lightning/utilities/cli.py b/pytorch_lightning/utilities/cli.py
index 214a41a16d407..7778915fac9fc 100644
--- a/pytorch_lightning/utilities/cli.py
+++ b/pytorch_lightning/utilities/cli.py
@@ -53,7 +53,7 @@ def add_lightning_class_args(
         self,
         lightning_class: Union['pl.Trainer', 'pl.LightningModule', 'pl.LightningDataModule'],
         nested_key: str,
-        subclass_mode: bool = False
+        subclass_mode: bool = False,
     ) -> None:
         """
         Adds arguments from a lightning class to a nested key of the parser
@@ -104,7 +104,7 @@ def __init__(
         env_parse: bool = False,
         parser_kwargs: Optional[Dict[str, Any]] = None,
         subclass_mode_model: bool = False,
-        subclass_mode_data: bool = False
+        subclass_mode_data: bool = False,
     ) -> None:
         """
         Receives as input pytorch-lightning classes, which are instantiated
diff --git a/pytorch_lightning/utilities/cloud_io.py b/pytorch_lightning/utilities/cloud_io.py
index d77c2f8745d31..93a328bba462b 100644
--- a/pytorch_lightning/utilities/cloud_io.py
+++ b/pytorch_lightning/utilities/cloud_io.py
@@ -24,7 +24,7 @@
 
 def load(
     path_or_url: Union[str, IO, Path],
-    map_location: Optional[Union[str, torch.device, Dict[Union[str, torch.device], Union[str, torch.device]]]] = None
+    map_location: Optional[Union[str, torch.device, Dict[Union[str, torch.device], Union[str, torch.device]]]] = None,
 ) -> Any:
     if not isinstance(path_or_url, (str, Path)):
         # any sort of BytesIO or similiar
diff --git a/pytorch_lightning/utilities/debugging.py b/pytorch_lightning/utilities/debugging.py
index d441b025388fe..3eeb28e2b5c3c 100644
--- a/pytorch_lightning/utilities/debugging.py
+++ b/pytorch_lightning/utilities/debugging.py
@@ -64,7 +64,7 @@ def track_event(
         evt_value: Optional[Any] = None,
         global_rank: Optional[int] = None,
         local_rank: Optional[int] = None,
-        comment: str = ''
+        comment: str = '',
     ) -> None:
         self.events.append({
             "timestamp": time.time(),
@@ -134,7 +134,7 @@ def track_lr_schedulers_update(
         old_lr: float,
         new_lr: float,
         monitor_key: Optional[str] = None,
-        monitor_val: Optional[torch.Tensor] = None
+        monitor_val: Optional[torch.Tensor] = None,
     ) -> None:
         loss_dict = {
             'batch_idx': batch_idx,
diff --git a/pytorch_lightning/utilities/distributed.py b/pytorch_lightning/utilities/distributed.py
index 7cee9efa2c052..1e7a89dcc27c9 100644
--- a/pytorch_lightning/utilities/distributed.py
+++ b/pytorch_lightning/utilities/distributed.py
@@ -116,7 +116,7 @@ def gather_all_tensors(result: torch.Tensor, group: Optional[Any] = None) -> Lis
 def sync_ddp_if_available(
     result: torch.Tensor,
     group: Optional[Any] = None,
-    reduce_op: Optional[Union[ReduceOp, str]] = None
+    reduce_op: Optional[Union[ReduceOp, str]] = None,
 ) -> torch.Tensor:
     """
     Function to reduce a tensor across worker processes during distributed training
@@ -137,7 +137,7 @@ def sync_ddp_if_available(
 def sync_ddp(
     result: torch.Tensor,
     group: Optional[Any] = None,
-    reduce_op: Optional[Union[ReduceOp, str]] = None
+    reduce_op: Optional[Union[ReduceOp, str]] = None,
 ) -> torch.Tensor:
     """
     Function to reduce the tensors from several ddp processes to one master process
diff --git a/pytorch_lightning/utilities/memory.py b/pytorch_lightning/utilities/memory.py
index 5658a28bd0036..27e4ca33022dd 100644
--- a/pytorch_lightning/utilities/memory.py
+++ b/pytorch_lightning/utilities/memory.py
@@ -20,7 +20,7 @@
 
 def recursive_detach(
     in_dict: Dict[Any, Union[torch.Tensor, Dict[Any, torch.Tensor]]],
-    to_cpu: bool = False
+    to_cpu: bool = False,
 ) -> Dict[str, torch.Tensor]:
     """Detach all tensors in `in_dict`.
 
diff --git a/pytorch_lightning/utilities/parsing.py b/pytorch_lightning/utilities/parsing.py
index 4331a149eb2d3..f11141d801284 100644
--- a/pytorch_lightning/utilities/parsing.py
+++ b/pytorch_lightning/utilities/parsing.py
@@ -188,7 +188,7 @@ def save_hyperparameters(
     obj: Any,
     *args: Any,
     ignore: Optional[Union[Sequence[str], str]] = None,
-    frame: Optional[types.FrameType] = None
+    frame: Optional[types.FrameType] = None,
 ) -> None:
     """See :meth:`~pytorch_lightning.LightningModule.save_hyperparameters`"""
 

From 99811aaacfd7691c3a9ce5c3a3e6f9575113a6a9 Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Thu, 6 May 2021 12:46:51 +0200
Subject: [PATCH 21/51] Add import Set to device_parser.py

---
 pytorch_lightning/utilities/device_parser.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pytorch_lightning/utilities/device_parser.py b/pytorch_lightning/utilities/device_parser.py
index 8e41900c9d196..64a781e7b9bac 100644
--- a/pytorch_lightning/utilities/device_parser.py
+++ b/pytorch_lightning/utilities/device_parser.py
@@ -12,7 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 import operator
-from typing import Any, List, MutableSequence, Optional, Tuple, Union
+from typing import Any, List, MutableSequence, Optional, Set, Tuple, Union
 
 import torch
 

From 7908c39ee12e21e4326214e6daaba4771ae42fcb Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Fri, 7 May 2021 21:41:05 +0200
Subject: [PATCH 22/51] Remove unused import to pass PEP8

---
 pytorch_lightning/utilities/apply_func.py | 1 -
 1 file changed, 1 deletion(-)

diff --git a/pytorch_lightning/utilities/apply_func.py b/pytorch_lightning/utilities/apply_func.py
index e000df7831b9b..c7cc3c3c173d8 100644
--- a/pytorch_lightning/utilities/apply_func.py
+++ b/pytorch_lightning/utilities/apply_func.py
@@ -21,7 +21,6 @@
 import numpy as np
 import torch
 
-from pytorch_lightning.utilities.exceptions import MisconfigurationException
 from pytorch_lightning.utilities.imports import _compare_version, _TORCHTEXT_AVAILABLE
 
 if _TORCHTEXT_AVAILABLE:

From 198aa413750e50568f27c922454d4e998d4ffe22 Mon Sep 17 00:00:00 2001
From: "pre-commit-ci[bot]"
 <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Date: Fri, 7 May 2021 19:42:00 +0000
Subject: [PATCH 23/51] [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci
---
 docs/source/governance.rst               | 2 --
 pytorch_lightning/utilities/argparse.py  | 2 +-
 pytorch_lightning/utilities/debugging.py | 2 +-
 3 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/docs/source/governance.rst b/docs/source/governance.rst
index fac8b68e1df53..5b1f9bd1916c1 100644
--- a/docs/source/governance.rst
+++ b/docs/source/governance.rst
@@ -38,5 +38,3 @@ Alumni
 - Jeff Ling (`jeffling <https://github.com/jeffling>`_)
 - Teddy Koker (`teddykoker <https://github.com/teddykoker>`_)
 - Nate Raw (`nateraw <https://github.com/nateraw>`_)
-
-
diff --git a/pytorch_lightning/utilities/argparse.py b/pytorch_lightning/utilities/argparse.py
index 75a8ff02ceb12..8f71084d77216 100644
--- a/pytorch_lightning/utilities/argparse.py
+++ b/pytorch_lightning/utilities/argparse.py
@@ -18,7 +18,7 @@
 from typing import Any, Dict, List, Tuple, Union
 
 import pytorch_lightning as pl
-from pytorch_lightning.utilities.parsing import str_to_bool, str_to_bool_or_str, str_to_bool_or_int
+from pytorch_lightning.utilities.parsing import str_to_bool, str_to_bool_or_int, str_to_bool_or_str
 
 
 def from_argparse_args(cls, args: Union[Namespace, ArgumentParser], **kwargs: Any) -> 'pl.Trainer':
diff --git a/pytorch_lightning/utilities/debugging.py b/pytorch_lightning/utilities/debugging.py
index 3eeb28e2b5c3c..3ff5b09e0bc11 100644
--- a/pytorch_lightning/utilities/debugging.py
+++ b/pytorch_lightning/utilities/debugging.py
@@ -16,7 +16,7 @@
 import time
 from collections import Counter
 from functools import wraps
-from typing import Any, Dict, Callable, List, Optional
+from typing import Any, Callable, Dict, List, Optional
 
 import torch
 from torch.utils.data import DataLoader

From 7fcbf96eff3c0ff7f3b8832c06f5b330013e657d Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Sat, 8 May 2021 10:45:52 +0200
Subject: [PATCH 24/51] Fix some issues after review

* Add Type to apply_func.py to emhpasize a class is expected to be passed in

* Replace torch.dtype with type in some functions in cli.py
---
 pytorch_lightning/utilities/apply_func.py | 2 +-
 pytorch_lightning/utilities/cli.py        | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/pytorch_lightning/utilities/apply_func.py b/pytorch_lightning/utilities/apply_func.py
index c7cc3c3c173d8..8675ed0eb9d32 100644
--- a/pytorch_lightning/utilities/apply_func.py
+++ b/pytorch_lightning/utilities/apply_func.py
@@ -55,7 +55,7 @@ def from_numpy(value: np.ndarray, device: Optional[Union[str, torch.device]] = N
 
 def apply_to_collection(
     data: Any,
-    dtype: Union[torch.dtype, Tuple[torch.dtype]],
+    dtype: Union[type, Tuple[type]],
     function: Callable,
     *args: Any,
     wrong_dtype: Optional[Union[torch.dtype, Tuple[torch.dtype]]] = None,
diff --git a/pytorch_lightning/utilities/cli.py b/pytorch_lightning/utilities/cli.py
index 7778915fac9fc..d7a916e1139ac 100644
--- a/pytorch_lightning/utilities/cli.py
+++ b/pytorch_lightning/utilities/cli.py
@@ -51,7 +51,7 @@ def __init__(self, *args: Any, parse_as_dict: bool = True, **kwargs: Any) -> Non
 
     def add_lightning_class_args(
         self,
-        lightning_class: Union['pl.Trainer', 'pl.LightningModule', 'pl.LightningDataModule'],
+        lightning_class: Union[Type('pl.Trainer'), Type('pl.LightningModule'), Type('pl.LightningDataModule')],
         nested_key: str,
         subclass_mode: bool = False,
     ) -> None:

From fef29070ab65f082a82a7bad780cbffecffcc04b Mon Sep 17 00:00:00 2001
From: "pre-commit-ci[bot]"
 <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Date: Sat, 8 May 2021 08:47:52 +0000
Subject: [PATCH 25/51] [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci
---
 pytorch_lightning/utilities/cli.py | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/pytorch_lightning/utilities/cli.py b/pytorch_lightning/utilities/cli.py
index d7a916e1139ac..033259c785258 100644
--- a/pytorch_lightning/utilities/cli.py
+++ b/pytorch_lightning/utilities/cli.py
@@ -51,7 +51,9 @@ def __init__(self, *args: Any, parse_as_dict: bool = True, **kwargs: Any) -> Non
 
     def add_lightning_class_args(
         self,
-        lightning_class: Union[Type('pl.Trainer'), Type('pl.LightningModule'), Type('pl.LightningDataModule')],
+        lightning_class: Union[Type('pl.Trainer'),
+                               Type('pl.LightningModule'),
+                               Type('pl.LightningDataModule')],
         nested_key: str,
         subclass_mode: bool = False,
     ) -> None:

From 60d615da7571b9b76382515501d868728570042c Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Sat, 8 May 2021 11:00:57 +0200
Subject: [PATCH 26/51] Fix typo a make consistent typing in cli.py

---
 pytorch_lightning/utilities/cli.py | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/pytorch_lightning/utilities/cli.py b/pytorch_lightning/utilities/cli.py
index 033259c785258..6db1e94889628 100644
--- a/pytorch_lightning/utilities/cli.py
+++ b/pytorch_lightning/utilities/cli.py
@@ -51,9 +51,9 @@ def __init__(self, *args: Any, parse_as_dict: bool = True, **kwargs: Any) -> Non
 
     def add_lightning_class_args(
         self,
-        lightning_class: Union[Type('pl.Trainer'),
-                               Type('pl.LightningModule'),
-                               Type('pl.LightningDataModule')],
+        lightning_class: Union[Type['pl.Trainer'],
+                               Type['pl.LightningModule'],
+                               Type['pl.LightningDataModule']],
         nested_key: str,
         subclass_mode: bool = False,
     ) -> None:
@@ -84,7 +84,7 @@ def __init__(
         self.config = config
         self.config_filename = config_filename
 
-    def on_train_start(self, trainer: Trainer, pl_module: LightningModule) -> None:
+    def on_train_start(self, trainer: 'pl.Trainer', pl_module: 'pl.LightningModule') -> None:
         log_dir = trainer.log_dir or trainer.default_root_dir
         config_path = os.path.join(log_dir, self.config_filename)
         self.parser.save(self.config, config_path, skip_none=False)

From 0854986da248cd39b17e3a2c8273e7dc3601c803 Mon Sep 17 00:00:00 2001
From: "pre-commit-ci[bot]"
 <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Date: Sat, 8 May 2021 09:01:45 +0000
Subject: [PATCH 27/51] [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci
---
 pytorch_lightning/utilities/cli.py | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/pytorch_lightning/utilities/cli.py b/pytorch_lightning/utilities/cli.py
index 6db1e94889628..c2c3a8f0bd7ae 100644
--- a/pytorch_lightning/utilities/cli.py
+++ b/pytorch_lightning/utilities/cli.py
@@ -51,9 +51,7 @@ def __init__(self, *args: Any, parse_as_dict: bool = True, **kwargs: Any) -> Non
 
     def add_lightning_class_args(
         self,
-        lightning_class: Union[Type['pl.Trainer'],
-                               Type['pl.LightningModule'],
-                               Type['pl.LightningDataModule']],
+        lightning_class: Union[Type['pl.Trainer'], Type['pl.LightningModule'], Type['pl.LightningDataModule']],
         nested_key: str,
         subclass_mode: bool = False,
     ) -> None:

From 14b172d93702f9a75523f0aac2e1bad03e43e469 Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Sat, 8 May 2021 11:10:26 +0200
Subject: [PATCH 28/51] * Reflect typing for recursive functions

* Introduce new typing class:
  recursive_dict_with_tensors = Dict[
    str, Union[torch.Tensor, 'recursive_dict_with_tensors']
]
---
 pytorch_lightning/utilities/memory.py  | 7 +++----
 pytorch_lightning/utilities/metrics.py | 4 +++-
 2 files changed, 6 insertions(+), 5 deletions(-)

diff --git a/pytorch_lightning/utilities/memory.py b/pytorch_lightning/utilities/memory.py
index 27e4ca33022dd..0273e86177ce3 100644
--- a/pytorch_lightning/utilities/memory.py
+++ b/pytorch_lightning/utilities/memory.py
@@ -17,11 +17,10 @@
 
 import torch
 
+recursive_dict_with_tensors = Dict[Any, Union[torch.Tensor, 'recursive_dict_with_tensors']]
 
-def recursive_detach(
-    in_dict: Dict[Any, Union[torch.Tensor, Dict[Any, torch.Tensor]]],
-    to_cpu: bool = False,
-) -> Dict[str, torch.Tensor]:
+
+def recursive_detach(in_dict: recursive_dict_with_tensors, to_cpu: bool = False) -> Dict[str, torch.Tensor]:
     """Detach all tensors in `in_dict`.
 
     May operate recursively if some of the values in `in_dict` are dictionaries
diff --git a/pytorch_lightning/utilities/metrics.py b/pytorch_lightning/utilities/metrics.py
index e547e69894a05..d1e7e6a23200e 100644
--- a/pytorch_lightning/utilities/metrics.py
+++ b/pytorch_lightning/utilities/metrics.py
@@ -19,8 +19,10 @@
 
 from pytorch_lightning.utilities.exceptions import MisconfigurationException
 
+recursive_dict_with_tensors = Dict[str, Union[torch.Tensor, 'recursive_dict_with_tensors']]
 
-def metrics_to_scalars(metrics: Dict[str, Union[torch.Tensor, dict]]) -> Dict[str, float]:
+
+def metrics_to_scalars(metrics: recursive_dict_with_tensors) -> Dict[str, float]:
     """ Recursively walk through a dictionary of metrics and convert single-item tensors to scalar values. """
 
     # TODO: this is duplicated in MetricsHolder. should be unified

From 6690bb9aecead774649c129188ffb88892f61331 Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Sat, 8 May 2021 12:56:19 +0200
Subject: [PATCH 29/51] Capitalize the name of recursive-dict type

---
 pytorch_lightning/utilities/memory.py  | 4 ++--
 pytorch_lightning/utilities/metrics.py | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/pytorch_lightning/utilities/memory.py b/pytorch_lightning/utilities/memory.py
index 0273e86177ce3..1a0dbcd04dff3 100644
--- a/pytorch_lightning/utilities/memory.py
+++ b/pytorch_lightning/utilities/memory.py
@@ -17,10 +17,10 @@
 
 import torch
 
-recursive_dict_with_tensors = Dict[Any, Union[torch.Tensor, 'recursive_dict_with_tensors']]
+RECURSIVE_DICT_WITH_TENSORS = Dict[Any, Union[torch.Tensor, 'RECURSIVE_DICT_WITH_TENSORS']]
 
 
-def recursive_detach(in_dict: recursive_dict_with_tensors, to_cpu: bool = False) -> Dict[str, torch.Tensor]:
+def recursive_detach(in_dict: RECURSIVE_DICT_WITH_TENSORS, to_cpu: bool = False) -> Dict[str, torch.Tensor]:
     """Detach all tensors in `in_dict`.
 
     May operate recursively if some of the values in `in_dict` are dictionaries
diff --git a/pytorch_lightning/utilities/metrics.py b/pytorch_lightning/utilities/metrics.py
index d1e7e6a23200e..1fd97eea0fc85 100644
--- a/pytorch_lightning/utilities/metrics.py
+++ b/pytorch_lightning/utilities/metrics.py
@@ -19,10 +19,10 @@
 
 from pytorch_lightning.utilities.exceptions import MisconfigurationException
 
-recursive_dict_with_tensors = Dict[str, Union[torch.Tensor, 'recursive_dict_with_tensors']]
+RECURSIVE_DICT_WITH_TENSORS = Dict[str, Union[torch.Tensor, 'RECURSIVE_DICT_WITH_TENSORS']]
 
 
-def metrics_to_scalars(metrics: recursive_dict_with_tensors) -> Dict[str, float]:
+def metrics_to_scalars(metrics: RECURSIVE_DICT_WITH_TENSORS) -> Dict[str, float]:
     """ Recursively walk through a dictionary of metrics and convert single-item tensors to scalar values. """
 
     # TODO: this is duplicated in MetricsHolder. should be unified

From 644776a530e6d81ff43c01a455e66337644b48fe Mon Sep 17 00:00:00 2001
From: jirka <jirka.borovec@seznam.cz>
Date: Tue, 11 May 2021 10:40:01 +0200
Subject: [PATCH 30/51] mypy

---
 setup.cfg | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/setup.cfg b/setup.cfg
index 3fa6e39076725..f3238d9b34cef 100644
--- a/setup.cfg
+++ b/setup.cfg
@@ -170,10 +170,6 @@ ignore_errors = True
 [mypy-pytorch_lightning.tuner.*]
 ignore_errors = True
 
-# todo: add proper typing to this module...
-[mypy-pytorch_lightning.utilities.*]
-ignore_errors = True
-
 # todo: add proper typing to this module...
 [mypy-pl_examples.*]
 ignore_errors = True

From e924379826df127a8a3e2162e7bd9f55a4d2712e Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Sat, 15 May 2021 11:50:56 +0200
Subject: [PATCH 31/51] Fix typing for enums.py and xla_device.py

---
 pytorch_lightning/utilities/enums.py      | 2 +-
 pytorch_lightning/utilities/xla_device.py | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/pytorch_lightning/utilities/enums.py b/pytorch_lightning/utilities/enums.py
index e01f8862486d3..766863819a97e 100644
--- a/pytorch_lightning/utilities/enums.py
+++ b/pytorch_lightning/utilities/enums.py
@@ -27,7 +27,7 @@ def from_str(cls, value: str) -> Optional['LightningEnum']:
                 return getattr(cls, st)
         return None
 
-    def __eq__(self, other: Union[str, Enum]) -> bool:
+    def __eq__(self, other: Union[object, Enum]) -> bool:
         other = other.value if isinstance(other, Enum) else str(other)
         return self.value.lower() == other.lower()
 
diff --git a/pytorch_lightning/utilities/xla_device.py b/pytorch_lightning/utilities/xla_device.py
index 5f82cff215ab5..91c21681308f1 100644
--- a/pytorch_lightning/utilities/xla_device.py
+++ b/pytorch_lightning/utilities/xla_device.py
@@ -40,7 +40,7 @@ def pl_multi_process(func: Callable) -> Callable:
 
     @functools.wraps(func)
     def wrapper(*args: Any, **kwargs: Any) -> Union[Any, bool]:
-        queue = Queue()
+        queue: Queue = Queue()
         proc = Process(target=inner_f, args=(queue, func, *args), kwargs=kwargs)
         proc.start()
         proc.join(TPU_CHECK_TIMEOUT)

From 2659a92035fff40dbba3ea3d2a65c969bab52588 Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Sat, 15 May 2021 12:09:45 +0200
Subject: [PATCH 32/51] Remove string types where not neede in 2 files

Files: cli.py, model_helpers.py
---
 pytorch_lightning/utilities/cli.py           | 10 +++++-----
 pytorch_lightning/utilities/model_helpers.py |  2 +-
 2 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/pytorch_lightning/utilities/cli.py b/pytorch_lightning/utilities/cli.py
index 2dcb23bd3a504..4a8f62795a977 100644
--- a/pytorch_lightning/utilities/cli.py
+++ b/pytorch_lightning/utilities/cli.py
@@ -51,7 +51,7 @@ def __init__(self, *args: Any, parse_as_dict: bool = True, **kwargs: Any) -> Non
 
     def add_lightning_class_args(
         self,
-        lightning_class: Union[Type['pl.Trainer'], Type['pl.LightningModule'], Type['pl.LightningDataModule']],
+        lightning_class: Union[Type[Trainer], Type[LightningModule], Type[LightningDataModule]],
         nested_key: str,
         subclass_mode: bool = False,
     ) -> None:
@@ -82,7 +82,7 @@ def __init__(
         self.config = config
         self.config_filename = config_filename
 
-    def on_train_start(self, trainer: 'pl.Trainer', pl_module: 'pl.LightningModule') -> None:
+    def on_train_start(self, trainer: Trainer, pl_module: LightningModule) -> None:
         log_dir = trainer.log_dir or trainer.default_root_dir
         config_path = os.path.join(log_dir, self.config_filename)
         self.parser.save(self.config, config_path, skip_none=False)
@@ -93,10 +93,10 @@ class LightningCLI:
 
     def __init__(
         self,
-        model_class: Type['pl.LightningModule'],
-        datamodule_class: Optional[Type['pl.LightningDataModule']] = None,
+        model_class: Type[LightningModule],
+        datamodule_class: Optional[Type[LightningDataModule]] = None,
         save_config_callback: Type[SaveConfigCallback] = SaveConfigCallback,
-        trainer_class: Type['pl.Trainer'] = Trainer,
+        trainer_class: Type[Trainer] = Trainer,
         trainer_defaults: Optional[Dict[str, Any]] = None,
         seed_everything_default: Optional[int] = None,
         description: str = 'pytorch-lightning trainer command line tool',
diff --git a/pytorch_lightning/utilities/model_helpers.py b/pytorch_lightning/utilities/model_helpers.py
index 6aaeb96dec599..88f0cc647c3b4 100644
--- a/pytorch_lightning/utilities/model_helpers.py
+++ b/pytorch_lightning/utilities/model_helpers.py
@@ -19,7 +19,7 @@
 from pytorch_lightning.core.lightning import LightningModule
 
 
-def is_overridden(method_name: str, model: Union['pl.LightningModule', 'pl.LightningDataModule']) -> bool:
+def is_overridden(method_name: str, model: Union[LightningModule, LightningDataModule]) -> bool:
     # if you pass DataModule instead of None or a LightningModule, we use LightningDataModule as super
     # TODO - refector this function to accept model_name, instance, parent so it makes more sense
     super_object = LightningModule if not isinstance(model, LightningDataModule) else LightningDataModule

From 0a16daeb651bd84ae6a0016722d74206206069ab Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Sat, 15 May 2021 12:53:51 +0200
Subject: [PATCH 33/51] Fix typing for utilities/argparse.py

---
 pytorch_lightning/utilities/argparse.py | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/pytorch_lightning/utilities/argparse.py b/pytorch_lightning/utilities/argparse.py
index 8f71084d77216..dcd940ee346fb 100644
--- a/pytorch_lightning/utilities/argparse.py
+++ b/pytorch_lightning/utilities/argparse.py
@@ -15,7 +15,7 @@
 import os
 from argparse import _ArgumentGroup, ArgumentParser, Namespace
 from contextlib import suppress
-from typing import Any, Dict, List, Tuple, Union
+from typing import Any, Callable, Dict, List, Tuple, Union
 
 import pytorch_lightning as pl
 from pytorch_lightning.utilities.parsing import str_to_bool, str_to_bool_or_int, str_to_bool_or_str
@@ -150,7 +150,7 @@ def add_argparse_args(
     parent_parser: ArgumentParser,
     *,
     use_argument_group: bool = True,
-) -> ArgumentParser:
+) -> Union[_ArgumentGroup, ArgumentParser]:
     r"""Extends existing argparse by default attributes for ``cls``.
 
     Args:
@@ -190,7 +190,7 @@ def add_argparse_args(
         raise RuntimeError("Please only pass an ArgumentParser instance.")
     if use_argument_group:
         group_name = _get_abbrev_qualified_cls_name(cls)
-        parser = parent_parser.add_argument_group(group_name)
+        parser: Union[_ArgumentGroup, ArgumentParser] = parent_parser.add_argument_group(group_name)
     else:
         parser = ArgumentParser(
             parents=[parent_parser],
@@ -213,16 +213,16 @@ def add_argparse_args(
     args_help = _parse_args_from_docstring(cls.__init__.__doc__ or cls.__doc__ or "")
 
     for arg, arg_types, arg_default in args_and_types:
-        arg_types = [at for at in allowed_types if at in arg_types]
+        arg_types = tuple(at for at in allowed_types if at in arg_types)
         if not arg_types:
             # skip argument with not supported type
             continue
-        arg_kwargs = {}
+        arg_kwargs: Dict[str, Any] = {}
         if bool in arg_types:
             arg_kwargs.update(nargs="?", const=True)
             # if the only arg type is bool
             if len(arg_types) == 1:
-                use_type = str_to_bool
+                use_type: Callable[[str], Union[bool, int, float, str]] = str_to_bool
             elif int in arg_types:
                 use_type = str_to_bool_or_int
             elif str in arg_types:
@@ -261,7 +261,7 @@ def add_argparse_args(
 
 def _parse_args_from_docstring(docstring: str) -> Dict[str, str]:
     arg_block_indent = None
-    current_arg = None
+    current_arg = ''
     parsed = {}
     for line in docstring.split("\n"):
         stripped = line.lstrip()

From db1a7d159132836ce0446a78f98a6f8bd980c155 Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Sat, 15 May 2021 13:40:34 +0200
Subject: [PATCH 34/51] Add missing typing for utilities/debugging.py

---
 pytorch_lightning/utilities/debugging.py | 38 ++++++++++++------------
 1 file changed, 19 insertions(+), 19 deletions(-)

diff --git a/pytorch_lightning/utilities/debugging.py b/pytorch_lightning/utilities/debugging.py
index 3ff5b09e0bc11..d390c899a3f67 100644
--- a/pytorch_lightning/utilities/debugging.py
+++ b/pytorch_lightning/utilities/debugging.py
@@ -16,7 +16,7 @@
 import time
 from collections import Counter
 from functools import wraps
-from typing import Any, Callable, Dict, List, Optional
+from typing import Any, Callable, Dict, List, Optional, Union
 
 import torch
 from torch.utils.data import DataLoader
@@ -44,19 +44,19 @@ class InternalDebugger(object):
     def __init__(self, trainer: 'pl.Trainer') -> None:
         self.enabled = os.environ.get('PL_DEV_DEBUG', '0') == '1'
         self.trainer = trainer
-        self.logged_metrics = []
-        self.pbar_added_metrics = []
-        self.saved_train_losses = []
-        self.saved_val_losses = []
-        self.saved_test_losses = []
-        self.early_stopping_history = []
-        self.checkpoint_callback_history = []
-        self.events = []
-        self.saved_lr_scheduler_updates = []
-        self.train_dataloader_calls = []
-        self.val_dataloader_calls = []
-        self.test_dataloader_calls = []
-        self.dataloader_sequence_calls = []
+        self.logged_metrics: List[Dict[str, Union[int, torch.Tensor]]] = []
+        self.pbar_added_metrics: List[Dict[str, Union[int, torch.Tensor]]] = []
+        self.saved_train_losses: List[Dict[str, Union[int, torch.Tensor, object]]] = []
+        self.saved_val_losses: List[Dict[str, Union[int, torch.Tensor, object]]] = []
+        self.saved_test_losses: List[Dict[str, Union[int, torch.Tensor, object]]] = []
+        self.early_stopping_history: List[Dict[str, Union[int, torch.Tensor, object]]] = []
+        self.checkpoint_callback_history: List[Dict[str, Union[int, torch.Tensor, object]]] = []
+        self.events: List[Dict[str, Any]] = []
+        self.saved_lr_scheduler_updates: List[Dict[str, Union[int, float, str, torch.Tensor, None]]] = []
+        self.train_dataloader_calls: List[Dict[str, Union[int, str, object]]] = []
+        self.val_dataloader_calls: List[Dict[str, Union[int, str, object]]] = []
+        self.test_dataloader_calls: List[Dict[str, Union[int, str, object]]] = []
+        self.dataloader_sequence_calls: List[Dict[str, Union[int, str, object]]] = []
 
     def track_event(
         self,
@@ -116,7 +116,7 @@ def track_load_dataloader_call(self, name: str, dataloaders: List[DataLoader]) -
             self.test_dataloader_calls.append(values)
 
     @enabled_only
-    def track_logged_metrics_history(self, scalar_metrics: Dict[str, torch.Tensor]) -> None:
+    def track_logged_metrics_history(self, scalar_metrics: Dict[str, Union[int, torch.Tensor]]) -> None:
         scalar_metrics['global_step'] = self.trainer.global_step
         self.logged_metrics.append(scalar_metrics)
 
@@ -164,12 +164,12 @@ def track_eval_loss_history(self, batch_idx: int, dataloader_idx: int, output: t
             self.saved_val_losses.append(loss_dict)
 
     @enabled_only
-    def track_pbar_metrics_history(self, metrics: Dict[str, torch.Tensor]) -> None:
+    def track_pbar_metrics_history(self, metrics: Dict[str, Union[int, torch.Tensor]]) -> None:
         metrics['debug_epoch'] = self.trainer.current_epoch
         self.pbar_added_metrics.append(metrics)
 
     @enabled_only
-    def track_early_stopping_history(self, callback: 'pl.Callback', current: torch.Tensor) -> None:
+    def track_early_stopping_history(self, callback: 'pl.callbacks.early_stopping.EarlyStopping', current: torch.Tensor) -> None:
         debug_dict = {
             'epoch': self.trainer.current_epoch,
             'global_step': self.trainer.global_step,
@@ -199,7 +199,7 @@ def num_seen_sanity_check_batches(self) -> int:
 
     @property
     def num_seen_val_check_batches(self) -> Counter:
-        counts = Counter()
+        counts: Counter = Counter()
         for x in self.saved_val_losses:
             if not x['sanity_check']:
                 counts.update({x['dataloader_idx']: 1})
@@ -207,7 +207,7 @@ def num_seen_val_check_batches(self) -> Counter:
 
     @property
     def num_seen_test_check_batches(self) -> Counter:
-        counts = Counter()
+        counts: Counter = Counter()
         for x in self.saved_test_losses:
             if not x['sanity_check']:
                 counts.update({x['dataloader_idx']: 1})

From 1993c79e6463095943357d3c177e03459e355b9b Mon Sep 17 00:00:00 2001
From: "pre-commit-ci[bot]"
 <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Date: Sat, 15 May 2021 11:41:36 +0000
Subject: [PATCH 35/51] [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci
---
 pytorch_lightning/utilities/debugging.py | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/pytorch_lightning/utilities/debugging.py b/pytorch_lightning/utilities/debugging.py
index d390c899a3f67..e6d8209b8a9d5 100644
--- a/pytorch_lightning/utilities/debugging.py
+++ b/pytorch_lightning/utilities/debugging.py
@@ -169,7 +169,9 @@ def track_pbar_metrics_history(self, metrics: Dict[str, Union[int, torch.Tensor]
         self.pbar_added_metrics.append(metrics)
 
     @enabled_only
-    def track_early_stopping_history(self, callback: 'pl.callbacks.early_stopping.EarlyStopping', current: torch.Tensor) -> None:
+    def track_early_stopping_history(
+        self, callback: 'pl.callbacks.early_stopping.EarlyStopping', current: torch.Tensor
+    ) -> None:
         debug_dict = {
             'epoch': self.trainer.current_epoch,
             'global_step': self.trainer.global_step,

From c6f2f4e332ba5f692b9fa5a693ce2743e56b55b3 Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Sat, 15 May 2021 14:20:39 +0200
Subject: [PATCH 36/51] Change type of NotImplemented to Any

---
 pytorch_lightning/utilities/apply_func.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pytorch_lightning/utilities/apply_func.py b/pytorch_lightning/utilities/apply_func.py
index 4db1e1546bfc1..dbdfa074627fc 100644
--- a/pytorch_lightning/utilities/apply_func.py
+++ b/pytorch_lightning/utilities/apply_func.py
@@ -122,7 +122,7 @@ class TransferableDataType(ABC):
     """
 
     @classmethod
-    def __subclasshook__(cls, subclass: Any) -> Union[bool, Type[NotImplemented]]:
+    def __subclasshook__(cls, subclass: Any) -> Union[bool, Any]:
         if cls is TransferableDataType:
             to = getattr(subclass, "to", None)
             return callable(to)

From 9cb30cc3edaac23cd9cb4e1449101a10ce14bc39 Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Sat, 15 May 2021 15:56:02 +0200
Subject: [PATCH 37/51] Fix mypy compatibility in a few files

* device_dtype_mixin.py - 0 issue

* device_parser.py - 0 issue

* [WIP] distributed.py - still some issues mainly related to rank_zero_only function
---
 .../utilities/device_dtype_mixin.py           | 21 +++++++-----
 pytorch_lightning/utilities/device_parser.py  | 21 +++++++-----
 pytorch_lightning/utilities/distributed.py    | 33 ++++++++++---------
 3 files changed, 44 insertions(+), 31 deletions(-)

diff --git a/pytorch_lightning/utilities/device_dtype_mixin.py b/pytorch_lightning/utilities/device_dtype_mixin.py
index c3a5d24461401..4f285d865c14e 100644
--- a/pytorch_lightning/utilities/device_dtype_mixin.py
+++ b/pytorch_lightning/utilities/device_dtype_mixin.py
@@ -114,7 +114,7 @@ def to(self, *args: Any, **kwargs: Any) -> Module:
         self.__update_properties(device=out[0], dtype=out[1])
         return super().to(*args, **kwargs)
 
-    def cuda(self, device: Optional[Union[torch.device, int]] = None) -> Module:
+    def cuda(self, device: Optional[Union[torch.device, int]] = None) -> 'DeviceDtypeModuleMixin':
         """Moves all model parameters and buffers to the GPU.
         This also makes associated parameters and buffers different objects. So
         it should be called before constructing optimizer if the module will
@@ -127,11 +127,16 @@ def cuda(self, device: Optional[Union[torch.device, int]] = None) -> Module:
         Returns:
             Module: self
         """
-        property_device = device if isinstance(device, torch.device) else torch.device('cuda', index=device)
+        if isinstance(device, torch.device):
+            property_device = device
+        elif isinstance(device, int):
+            property_device = torch.device('cuda', index=device)
+        else:
+            property_device = torch.device('cuda')  # use current cuda device if device == None [mypy compatibility]
         self.__update_properties(device=property_device)
         return super().cuda(device=device)
 
-    def cpu(self) -> Module:
+    def cpu(self) -> 'DeviceDtypeModuleMixin':
         """Moves all model parameters and buffers to the CPU.
 
         Returns:
@@ -140,7 +145,7 @@ def cpu(self) -> Module:
         self.__update_properties(device=torch.device('cpu'))
         return super().cpu()
 
-    def type(self, dst_type: Union[str, torch.dtype]) -> Module:
+    def type(self, dst_type: Union[str, torch.dtype]) -> 'DeviceDtypeModuleMixin':
         """Casts all parameters and buffers to :attr:`dst_type`.
 
         Arguments:
@@ -152,7 +157,7 @@ def type(self, dst_type: Union[str, torch.dtype]) -> Module:
         self.__update_properties(dtype=dst_type)
         return super().type(dst_type=dst_type)
 
-    def float(self) -> Module:
+    def float(self) -> 'DeviceDtypeModuleMixin':
         """Casts all floating point parameters and buffers to float datatype.
 
         Returns:
@@ -161,7 +166,7 @@ def float(self) -> Module:
         self.__update_properties(dtype=torch.float)
         return super().float()
 
-    def double(self) -> Module:
+    def double(self) -> 'DeviceDtypeModuleMixin':
         """Casts all floating point parameters and buffers to ``double`` datatype.
 
         Returns:
@@ -170,7 +175,7 @@ def double(self) -> Module:
         self.__update_properties(dtype=torch.double)
         return super().double()
 
-    def half(self) -> Module:
+    def half(self) -> 'DeviceDtypeModuleMixin':
         """Casts all floating point parameters and buffers to ``half`` datatype.
 
         Returns:
@@ -179,7 +184,7 @@ def half(self) -> Module:
         self.__update_properties(dtype=torch.half)
         return super().half()
 
-    def __update_properties(self, device: Optional[torch.device] = None, dtype: Optional[torch.dtype] = None) -> None:
+    def __update_properties(self, device: Optional[torch.device] = None, dtype: Optional[Union[str, torch.dtype]] = None) -> None:
 
         def apply_fn(module):
             if not isinstance(module, DeviceDtypeModuleMixin):
diff --git a/pytorch_lightning/utilities/device_parser.py b/pytorch_lightning/utilities/device_parser.py
index 64a781e7b9bac..db60216cd5928 100644
--- a/pytorch_lightning/utilities/device_parser.py
+++ b/pytorch_lightning/utilities/device_parser.py
@@ -12,7 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 import operator
-from typing import Any, List, MutableSequence, Optional, Set, Tuple, Union
+from typing import Any, Iterable, List, MutableSequence, Optional, Set, Tuple, Union
 
 import torch
 
@@ -187,15 +187,20 @@ def _check_data_type(device_ids: Any) -> None:
         raise MisconfigurationException("Device ID's (GPU/TPU) must be int, string or sequence of ints or None.")
 
 
-def _tpu_cores_valid(tpu_cores: Optional[Union[int, List[int], Tuple[int], Set[int]]]) -> bool:
+def _tpu_cores_valid(tpu_cores: Optional[Union[int, Iterable[int], List[int], Tuple[int], Set[int]]] = None) -> bool:
     # allow 1 or 8 cores
     if tpu_cores in (1, 8, None):
         return True
+    
+    # First condition is necessary for mypy compatibility;
+    # list_tpu_cores is required to declare for mypy compatiblity too
+    if (tpu_cores is not None) and (not isinstance(tpu_cores, int)):
+        list_tpu_cores: List[int] = list(tpu_cores)
 
     # allow picking 1 of 8 indexes
-    if isinstance(tpu_cores, (List, Tuple, Set)):
-        has_1_tpu_idx = len(tpu_cores) == 1
-        is_valid_tpu_idx = tpu_cores[0] in range(1, 9)
+    if isinstance(list_tpu_cores, List):
+        has_1_tpu_idx = len(list_tpu_cores) == 1
+        is_valid_tpu_idx = list_tpu_cores[0] in range(1, 9)
 
         is_valid_tpu_core_choice = has_1_tpu_idx and is_valid_tpu_idx
         return is_valid_tpu_core_choice
@@ -205,7 +210,7 @@ def _tpu_cores_valid(tpu_cores: Optional[Union[int, List[int], Tuple[int], Set[i
 
 def _parse_tpu_cores_str(tpu_cores: str) -> Union[int, List[int]]:
     if tpu_cores in ('1', '8'):
-        tpu_cores = int(tpu_cores)
+        int_tpu_cores: Union[int, List[int]] = int(tpu_cores)
     else:
-        tpu_cores = [int(x.strip()) for x in tpu_cores.split(',') if len(x) > 0]
-    return tpu_cores
+        int_tpu_cores = [int(x.strip()) for x in tpu_cores.split(',') if len(x) > 0]
+    return int_tpu_cores
diff --git a/pytorch_lightning/utilities/distributed.py b/pytorch_lightning/utilities/distributed.py
index 1e7a89dcc27c9..45c8e760d4d6e 100644
--- a/pytorch_lightning/utilities/distributed.py
+++ b/pytorch_lightning/utilities/distributed.py
@@ -16,7 +16,7 @@
 import os
 import warnings
 from functools import partial, wraps
-from typing import Any, Callable, List, Optional, Sequence, Tuple, Union
+from typing import Any, Callable, List, Optional, Tuple, Union
 
 import torch
 from torch.nn.parallel.distributed import DistributedDataParallel
@@ -41,12 +41,13 @@ class group:
 log = logging.getLogger(__name__)
 
 
-def rank_zero_only(fn: Callable) -> Optional[Callable]:
+def rank_zero_only(fn: Callable) -> Callable:
 
     @wraps(fn)
     def wrapped_fn(*args: Any, **kwargs: Any) -> Optional[Any]:
         if rank_zero_only.rank == 0:
             return fn(*args, **kwargs)
+        return None
 
     return wrapped_fn
 
@@ -174,23 +175,23 @@ def sync_ddp(
 class AllGatherGrad(torch.autograd.Function):
 
     @staticmethod
-    def forward(ctx: object, tensor: torch.Tensor, group=group.WORLD) -> torch.Tensor:
+    def forward(ctx: torch.autograd.Function, tensor: torch.Tensor, group=group.WORLD) -> torch.Tensor:
         ctx.group = group
 
         gathered_tensor = [torch.zeros_like(tensor) for _ in range(torch.distributed.get_world_size())]
 
         torch.distributed.all_gather(gathered_tensor, tensor, group=group)
-        gathered_tensor = torch.stack(gathered_tensor, dim=0)
+        stacked_tensor = torch.stack(gathered_tensor, dim=0)
 
-        return gathered_tensor
+        return stacked_tensor
 
     @staticmethod
-    def backward(ctx: object, *grad_output: Sequence[torch.Tensor]) -> Tuple[torch.Tensor, None]:
-        grad_output = torch.cat(grad_output)
+    def backward(ctx: torch.autograd.Function, *grad_output: torch.Tensor) -> Tuple[torch.Tensor, None]:
+        grad_output_cat = torch.cat(grad_output)
 
-        torch.distributed.all_reduce(grad_output, op=torch.distributed.ReduceOp.SUM, async_op=False, group=ctx.group)
+        torch.distributed.all_reduce(grad_output_cat, op=torch.distributed.ReduceOp.SUM, async_op=False, group=ctx.group)
 
-        return grad_output[torch.distributed.get_rank()], None
+        return grad_output_cat[torch.distributed.get_rank()], None
 
 
 def all_gather_ddp_if_available(
@@ -307,12 +308,14 @@ def register_ddp_comm_hook(
                 f"DDP comm wrapper is provided, apply {ddp_comm_wrapper.__qualname__}({ddp_comm_hook.__qualname__})."
             )
             ddp_comm_hook = ddp_comm_wrapper(ddp_comm_hook)
-
-    rank_zero_debug(f"Registering DDP comm hook: {ddp_comm_hook.__qualname__}.")
-    model.register_comm_hook(
-        state=ddp_comm_state,
-        hook=ddp_comm_hook,
-    )
+    
+    # If condition required for mypy compatibility
+    if ddp_comm_hook is not None:
+        rank_zero_debug(f"Registering DDP comm hook: {ddp_comm_hook.__qualname__}.")
+        model.register_comm_hook(
+            state=ddp_comm_state,
+            hook=ddp_comm_hook,
+        )
 
 
 def tpu_distributed() -> bool:

From 56301122bb9367481e73edad8fe08c4f8871a785 Mon Sep 17 00:00:00 2001
From: "pre-commit-ci[bot]"
 <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Date: Sat, 15 May 2021 13:59:38 +0000
Subject: [PATCH 38/51] [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci
---
 pytorch_lightning/utilities/device_dtype_mixin.py | 4 +++-
 pytorch_lightning/utilities/device_parser.py      | 2 +-
 pytorch_lightning/utilities/distributed.py        | 6 ++++--
 3 files changed, 8 insertions(+), 4 deletions(-)

diff --git a/pytorch_lightning/utilities/device_dtype_mixin.py b/pytorch_lightning/utilities/device_dtype_mixin.py
index 4f285d865c14e..bfecd7905a00c 100644
--- a/pytorch_lightning/utilities/device_dtype_mixin.py
+++ b/pytorch_lightning/utilities/device_dtype_mixin.py
@@ -184,7 +184,9 @@ def half(self) -> 'DeviceDtypeModuleMixin':
         self.__update_properties(dtype=torch.half)
         return super().half()
 
-    def __update_properties(self, device: Optional[torch.device] = None, dtype: Optional[Union[str, torch.dtype]] = None) -> None:
+    def __update_properties(
+        self, device: Optional[torch.device] = None, dtype: Optional[Union[str, torch.dtype]] = None
+    ) -> None:
 
         def apply_fn(module):
             if not isinstance(module, DeviceDtypeModuleMixin):
diff --git a/pytorch_lightning/utilities/device_parser.py b/pytorch_lightning/utilities/device_parser.py
index db60216cd5928..7111e1be7c288 100644
--- a/pytorch_lightning/utilities/device_parser.py
+++ b/pytorch_lightning/utilities/device_parser.py
@@ -191,7 +191,7 @@ def _tpu_cores_valid(tpu_cores: Optional[Union[int, Iterable[int], List[int], Tu
     # allow 1 or 8 cores
     if tpu_cores in (1, 8, None):
         return True
-    
+
     # First condition is necessary for mypy compatibility;
     # list_tpu_cores is required to declare for mypy compatiblity too
     if (tpu_cores is not None) and (not isinstance(tpu_cores, int)):
diff --git a/pytorch_lightning/utilities/distributed.py b/pytorch_lightning/utilities/distributed.py
index 45c8e760d4d6e..28723bbbfa291 100644
--- a/pytorch_lightning/utilities/distributed.py
+++ b/pytorch_lightning/utilities/distributed.py
@@ -189,7 +189,9 @@ def forward(ctx: torch.autograd.Function, tensor: torch.Tensor, group=group.WORL
     def backward(ctx: torch.autograd.Function, *grad_output: torch.Tensor) -> Tuple[torch.Tensor, None]:
         grad_output_cat = torch.cat(grad_output)
 
-        torch.distributed.all_reduce(grad_output_cat, op=torch.distributed.ReduceOp.SUM, async_op=False, group=ctx.group)
+        torch.distributed.all_reduce(
+            grad_output_cat, op=torch.distributed.ReduceOp.SUM, async_op=False, group=ctx.group
+        )
 
         return grad_output_cat[torch.distributed.get_rank()], None
 
@@ -308,7 +310,7 @@ def register_ddp_comm_hook(
                 f"DDP comm wrapper is provided, apply {ddp_comm_wrapper.__qualname__}({ddp_comm_hook.__qualname__})."
             )
             ddp_comm_hook = ddp_comm_wrapper(ddp_comm_hook)
-    
+
     # If condition required for mypy compatibility
     if ddp_comm_hook is not None:
         rank_zero_debug(f"Registering DDP comm hook: {ddp_comm_hook.__qualname__}.")

From 00a145b0fa7b5d0defe89e337d838d639dfb4673 Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Sat, 15 May 2021 16:52:36 +0200
Subject: [PATCH 39/51] [WIP] Fix mypy compatibility for parsing.py

* TODO: Resolve one missing return and non-FrameType in save_hyperparameters function.
---
 pytorch_lightning/utilities/parsing.py | 51 +++++++++++++++-----------
 1 file changed, 30 insertions(+), 21 deletions(-)

diff --git a/pytorch_lightning/utilities/parsing.py b/pytorch_lightning/utilities/parsing.py
index f11141d801284..747c22bae7fc2 100644
--- a/pytorch_lightning/utilities/parsing.py
+++ b/pytorch_lightning/utilities/parsing.py
@@ -12,11 +12,14 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 import copy
+from enum import Enum
 import inspect
 import pickle
+from pytorch_lightning.core.lightning import LightningModule
 import types
 from argparse import Namespace
-from typing import Any, Dict, List, Optional, Sequence, Tuple, Type, Union
+from types import FrameType
+from typing import Any, Dict, List, Literal, Optional, Sequence, Tuple, Union
 
 import pytorch_lightning as pl
 from pytorch_lightning.utilities import rank_zero_warn
@@ -51,10 +54,10 @@ def str_to_bool(val: str) -> bool:
     >>> str_to_bool('FALSE')
     False
     """
-    val = str_to_bool_or_str(val)
-    if isinstance(val, bool):
-        return val
-    raise ValueError(f'invalid truth value {val}')
+    val_converted = str_to_bool_or_str(val)
+    if isinstance(val_converted, bool):
+        return val_converted
+    raise ValueError(f'invalid truth value {val_converted}')
 
 
 def str_to_bool_or_int(val: str) -> Union[bool, int, str]:
@@ -69,13 +72,13 @@ def str_to_bool_or_int(val: str) -> Union[bool, int, str]:
     >>> str_to_bool_or_int("abc")
     'abc'
     """
-    val = str_to_bool_or_str(val)
-    if isinstance(val, bool):
-        return val
+    val_converted = str_to_bool_or_str(val)
+    if isinstance(val_converted, bool):
+        return val_converted
     try:
-        return int(val)
+        return int(val_converted)
     except ValueError:
-        return val
+        return val_converted
 
 
 def is_picklable(obj: object) -> bool:
@@ -118,7 +121,10 @@ def parse_class_init_keys(cls) -> Tuple[str, str, str]:
     # self is always first
     n_self = init_params[0].name
 
-    def _get_first_if_any(params: Sequence[Namespace], param_type: Type) -> str:
+    def _get_first_if_any(
+        params: List[inspect.Parameter],
+        param_type: Literal[inspect._ParameterKind.VAR_POSITIONAL, inspect._ParameterKind.VAR_KEYWORD],
+    ) -> str:
         for p in params:
             if p.kind == param_type:
                 return p.name
@@ -129,7 +135,7 @@ def _get_first_if_any(params: Sequence[Namespace], param_type: Type) -> str:
     return n_self, n_args, n_kwargs
 
 
-def get_init_args(frame: object) -> Dict[str, Any]:
+def get_init_args(frame: FrameType) -> Dict[str, Any]:
     _, _, _, local_vars = inspect.getargvalues(frame)
     if '__class__' not in local_vars:
         return {}
@@ -145,7 +151,7 @@ def get_init_args(frame: object) -> Dict[str, Any]:
     return local_args
 
 
-def collect_init_args(frame: object, path_args: List[Dict[str, Any]], inside: bool = False) -> List[Dict[str, Any]]:
+def collect_init_args(frame: FrameType, path_args: List[Dict[str, Any]], inside: bool = False) -> List[Dict[str, Any]]:
     """
     Recursively collects the arguments passed to the child constructors in the inheritance tree.
 
@@ -160,13 +166,16 @@ def collect_init_args(frame: object, path_args: List[Dict[str, Any]], inside: bo
           most specific class in the hierarchy.
     """
     _, _, _, local_vars = inspect.getargvalues(frame)
-    if '__class__' in local_vars:
-        local_args = get_init_args(frame)
-        # recursive update
-        path_args.append(local_args)
-        return collect_init_args(frame.f_back, path_args, inside=True)
-    elif not inside:
-        return collect_init_args(frame.f_back, path_args, inside)
+    if isinstance(frame.f_back, FrameType):
+        if '__class__' in local_vars:
+            local_args = get_init_args(frame)
+            # recursive update
+            path_args.append(local_args)
+            return collect_init_args(frame.f_back, path_args, inside=True)
+        elif not inside:
+            return collect_init_args(frame.f_back, path_args, inside)
+        else:
+            return path_args
     else:
         return path_args
 
@@ -272,7 +281,7 @@ def _lightning_get_all_attr_holders(model: 'pl.LightningModule', attribute: str)
     """
     trainer = getattr(model, 'trainer', None)
 
-    holders = []
+    holders: List[Union[Dict[Any, Any], Namespace, LightningModule]] = []
 
     # Check if attribute in model
     if hasattr(model, attribute):

From 87be8a08e9da2d1faf216f6993c312db9f211916 Mon Sep 17 00:00:00 2001
From: "pre-commit-ci[bot]"
 <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Date: Sat, 15 May 2021 14:55:27 +0000
Subject: [PATCH 40/51] [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci
---
 pytorch_lightning/utilities/parsing.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/pytorch_lightning/utilities/parsing.py b/pytorch_lightning/utilities/parsing.py
index 747c22bae7fc2..2a8153f42b5ec 100644
--- a/pytorch_lightning/utilities/parsing.py
+++ b/pytorch_lightning/utilities/parsing.py
@@ -12,16 +12,16 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 import copy
-from enum import Enum
 import inspect
 import pickle
-from pytorch_lightning.core.lightning import LightningModule
 import types
 from argparse import Namespace
+from enum import Enum
 from types import FrameType
 from typing import Any, Dict, List, Literal, Optional, Sequence, Tuple, Union
 
 import pytorch_lightning as pl
+from pytorch_lightning.core.lightning import LightningModule
 from pytorch_lightning.utilities import rank_zero_warn
 
 

From 8eda497d121d64191883ae270d34a193e0777a83 Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Sat, 15 May 2021 17:23:40 +0200
Subject: [PATCH 41/51] Import Literal from typing_extensions to support python
 version < 3.8

---
 pytorch_lightning/utilities/parsing.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/pytorch_lightning/utilities/parsing.py b/pytorch_lightning/utilities/parsing.py
index 2a8153f42b5ec..733204b64dc72 100644
--- a/pytorch_lightning/utilities/parsing.py
+++ b/pytorch_lightning/utilities/parsing.py
@@ -18,7 +18,8 @@
 from argparse import Namespace
 from enum import Enum
 from types import FrameType
-from typing import Any, Dict, List, Literal, Optional, Sequence, Tuple, Union
+from typing import Any, Dict, List, Optional, Sequence, Tuple, Union
+from typing_extensions import Literal
 
 import pytorch_lightning as pl
 from pytorch_lightning.core.lightning import LightningModule

From d198abf07b9c4c5ad9abca83bf6a9ae859d4c5a8 Mon Sep 17 00:00:00 2001
From: "pre-commit-ci[bot]"
 <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Date: Sat, 15 May 2021 15:24:37 +0000
Subject: [PATCH 42/51] [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci
---
 pytorch_lightning/utilities/parsing.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/pytorch_lightning/utilities/parsing.py b/pytorch_lightning/utilities/parsing.py
index 733204b64dc72..336f3b433b611 100644
--- a/pytorch_lightning/utilities/parsing.py
+++ b/pytorch_lightning/utilities/parsing.py
@@ -19,6 +19,7 @@
 from enum import Enum
 from types import FrameType
 from typing import Any, Dict, List, Optional, Sequence, Tuple, Union
+
 from typing_extensions import Literal
 
 import pytorch_lightning as pl

From f1e0e7edd9262a7fb0d283d72d9c6a919571dc59 Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Sat, 15 May 2021 17:47:07 +0200
Subject: [PATCH 43/51] Remove unusued import and circular import

---
 pytorch_lightning/utilities/apply_func.py    |  2 +-
 pytorch_lightning/utilities/cli.py           |  1 -
 pytorch_lightning/utilities/model_helpers.py |  1 -
 pytorch_lightning/utilities/parsing.py       | 16 +++++++---------
 4 files changed, 8 insertions(+), 12 deletions(-)

diff --git a/pytorch_lightning/utilities/apply_func.py b/pytorch_lightning/utilities/apply_func.py
index dbdfa074627fc..29d0bdf935466 100644
--- a/pytorch_lightning/utilities/apply_func.py
+++ b/pytorch_lightning/utilities/apply_func.py
@@ -16,7 +16,7 @@
 from collections.abc import Mapping, Sequence
 from copy import copy
 from functools import partial
-from typing import Any, Callable, List, Optional, Tuple, Type, Union
+from typing import Any, Callable, List, Optional, Tuple, Union
 
 import numpy as np
 import torch
diff --git a/pytorch_lightning/utilities/cli.py b/pytorch_lightning/utilities/cli.py
index 4a8f62795a977..94b969d713f3a 100644
--- a/pytorch_lightning/utilities/cli.py
+++ b/pytorch_lightning/utilities/cli.py
@@ -15,7 +15,6 @@
 from argparse import Namespace
 from typing import Any, Dict, Optional, Type, Union
 
-import pytorch_lightning as pl
 from pytorch_lightning.callbacks import Callback
 from pytorch_lightning.core.datamodule import LightningDataModule
 from pytorch_lightning.core.lightning import LightningModule
diff --git a/pytorch_lightning/utilities/model_helpers.py b/pytorch_lightning/utilities/model_helpers.py
index 88f0cc647c3b4..87bd9e6c4545d 100644
--- a/pytorch_lightning/utilities/model_helpers.py
+++ b/pytorch_lightning/utilities/model_helpers.py
@@ -14,7 +14,6 @@
 
 from typing import Union
 
-import pytorch_lightning as pl
 from pytorch_lightning.core.datamodule import LightningDataModule
 from pytorch_lightning.core.lightning import LightningModule
 
diff --git a/pytorch_lightning/utilities/parsing.py b/pytorch_lightning/utilities/parsing.py
index 336f3b433b611..fc7b24a6b286b 100644
--- a/pytorch_lightning/utilities/parsing.py
+++ b/pytorch_lightning/utilities/parsing.py
@@ -16,14 +16,12 @@
 import pickle
 import types
 from argparse import Namespace
-from enum import Enum
 from types import FrameType
 from typing import Any, Dict, List, Optional, Sequence, Tuple, Union
 
 from typing_extensions import Literal
 
 import pytorch_lightning as pl
-from pytorch_lightning.core.lightning import LightningModule
 from pytorch_lightning.utilities import rank_zero_warn
 
 
@@ -201,7 +199,7 @@ def save_hyperparameters(
     ignore: Optional[Union[Sequence[str], str]] = None,
     frame: Optional[types.FrameType] = None,
 ) -> None:
-    """See :meth:`~pytorch_lightning.LightningModule.save_hyperparameters`"""
+    """See :meth:`~pytorch_lightning.'pl.core.lightning.LightningModule'.save_hyperparameters`"""
 
     if len(args) == 1 and not isinstance(args, str) and not args[0]:
         # args[0] is an empty container
@@ -276,14 +274,14 @@ def __repr__(self) -> str:
         return out
 
 
-def _lightning_get_all_attr_holders(model: 'pl.LightningModule', attribute: str) -> Any:
+def _lightning_get_all_attr_holders(model: 'pl.core.lightning.LightningModule', attribute: str) -> Any:
     """
     Special attribute finding for Lightning. Gets all of the objects or dicts that holds attribute.
     Checks for attribute in model namespace, the old hparams namespace/dict, and the datamodule.
     """
     trainer = getattr(model, 'trainer', None)
 
-    holders: List[Union[Dict[Any, Any], Namespace, LightningModule]] = []
+    holders: List[Union[Dict[Any, Any], Namespace, 'pl.core.lightning.LightningModule']] = []
 
     # Check if attribute in model
     if hasattr(model, attribute):
@@ -301,7 +299,7 @@ def _lightning_get_all_attr_holders(model: 'pl.LightningModule', attribute: str)
     return holders
 
 
-def _lightning_get_first_attr_holder(model: 'pl.LightningModule', attribute: str) -> Optional[Any]:
+def _lightning_get_first_attr_holder(model: 'pl.core.lightning.LightningModule', attribute: str) -> Optional[Any]:
     """
     Special attribute finding for Lightning.  Gets the object or dict that holds attribute, or None.
     Checks for attribute in model namespace, the old hparams namespace/dict, and the datamodule,
@@ -314,7 +312,7 @@ def _lightning_get_first_attr_holder(model: 'pl.LightningModule', attribute: str
     return holders[-1]
 
 
-def lightning_hasattr(model: 'pl.LightningModule', attribute: str) -> bool:
+def lightning_hasattr(model: 'pl.core.lightning.LightningModule', attribute: str) -> bool:
     """
     Special hasattr for Lightning. Checks for attribute in model namespace,
     the old hparams namespace/dict, and the datamodule.
@@ -322,7 +320,7 @@ def lightning_hasattr(model: 'pl.LightningModule', attribute: str) -> bool:
     return _lightning_get_first_attr_holder(model, attribute) is not None
 
 
-def lightning_getattr(model: 'pl.LightningModule', attribute: str) -> Any:
+def lightning_getattr(model: 'pl.core.lightning.LightningModule', attribute: str) -> Any:
     """
     Special getattr for Lightning. Checks for attribute in model namespace,
     the old hparams namespace/dict, and the datamodule.
@@ -344,7 +342,7 @@ def lightning_getattr(model: 'pl.LightningModule', attribute: str) -> Any:
     return getattr(holder, attribute)
 
 
-def lightning_setattr(model: 'pl.LightningModule', attribute: str, value: Any) -> None:
+def lightning_setattr(model: 'pl.core.lightning.LightningModule', attribute: str, value: Any) -> None:
     """
     Special setattr for Lightning. Checks for attribute in model namespace
     and the old hparams namespace/dict.

From 94f7bf74daf6094765e5170549de72ca91abced0 Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Sat, 15 May 2021 20:36:38 +0200
Subject: [PATCH 44/51] Fix another bunch of mypy issues

---
 pytorch_lightning/utilities/argparse.py           | 14 +++++++-------
 pytorch_lightning/utilities/device_dtype_mixin.py |  4 ++--
 pytorch_lightning/utilities/distributed.py        | 10 +++++++---
 pytorch_lightning/utilities/memory.py             |  6 ++++--
 pytorch_lightning/utilities/metrics.py            |  6 +++---
 pytorch_lightning/utilities/parsing.py            |  4 ++--
 pytorch_lightning/utilities/warnings.py           |  4 ++--
 7 files changed, 27 insertions(+), 21 deletions(-)

diff --git a/pytorch_lightning/utilities/argparse.py b/pytorch_lightning/utilities/argparse.py
index dcd940ee346fb..b4dd145b9da7b 100644
--- a/pytorch_lightning/utilities/argparse.py
+++ b/pytorch_lightning/utilities/argparse.py
@@ -15,13 +15,13 @@
 import os
 from argparse import _ArgumentGroup, ArgumentParser, Namespace
 from contextlib import suppress
-from typing import Any, Callable, Dict, List, Tuple, Union
+from typing import Any, Callable, Dict, List, Tuple, Type, Union
 
 import pytorch_lightning as pl
 from pytorch_lightning.utilities.parsing import str_to_bool, str_to_bool_or_int, str_to_bool_or_str
 
 
-def from_argparse_args(cls, args: Union[Namespace, ArgumentParser], **kwargs: Any) -> 'pl.Trainer':
+def from_argparse_args(cls: Type['pl.Trainer'], args: Union[Namespace, ArgumentParser], **kwargs: Any) -> 'pl.Trainer':
     """Create an instance from CLI arguments.
     Eventually use varibles from OS environement which are defined as "PL_<CLASS-NAME>_<CLASS_ARUMENT_NAME>"
 
@@ -53,7 +53,7 @@ def from_argparse_args(cls, args: Union[Namespace, ArgumentParser], **kwargs: An
     return cls(**trainer_kwargs)
 
 
-def parse_argparser(cls, arg_parser: Union[ArgumentParser, Namespace]) -> Namespace:
+def parse_argparser(cls: Type['pl.Trainer'], arg_parser: Union[ArgumentParser, Namespace]) -> Namespace:
     """Parse CLI arguments, required for custom bool types."""
     args = arg_parser.parse_args() if isinstance(arg_parser, ArgumentParser) else arg_parser
 
@@ -78,7 +78,7 @@ def parse_argparser(cls, arg_parser: Union[ArgumentParser, Namespace]) -> Namesp
     return Namespace(**modified_args)
 
 
-def parse_env_variables(cls, template: str = "PL_%(cls_name)s_%(cls_argument)s") -> Namespace:
+def parse_env_variables(cls: Type['pl.Trainer'], template: str = "PL_%(cls_name)s_%(cls_argument)s") -> Namespace:
     """Parse environment arguments if they are defined.
 
     Example:
@@ -107,7 +107,7 @@ def parse_env_variables(cls, template: str = "PL_%(cls_name)s_%(cls_argument)s")
     return Namespace(**env_args)
 
 
-def get_init_arguments_and_types(cls) -> List[Tuple[str, Tuple, Any]]:
+def get_init_arguments_and_types(cls: Type['pl.Trainer']) -> List[Tuple[str, Tuple, Any]]:
     r"""Scans the class signature and returns argument names, types and default values.
 
     Returns:
@@ -135,7 +135,7 @@ def get_init_arguments_and_types(cls) -> List[Tuple[str, Tuple, Any]]:
     return name_type_default
 
 
-def _get_abbrev_qualified_cls_name(cls) -> str:
+def _get_abbrev_qualified_cls_name(cls: Type['pl.Trainer']) -> str:
     assert isinstance(cls, type), repr(cls)
     if cls.__module__.startswith("pytorch_lightning."):
         # Abbreviate.
@@ -146,7 +146,7 @@ def _get_abbrev_qualified_cls_name(cls) -> str:
 
 
 def add_argparse_args(
-    cls,
+    cls: Type['pl.Trainer'],
     parent_parser: ArgumentParser,
     *,
     use_argument_group: bool = True,
diff --git a/pytorch_lightning/utilities/device_dtype_mixin.py b/pytorch_lightning/utilities/device_dtype_mixin.py
index bfecd7905a00c..6a357a27622dd 100644
--- a/pytorch_lightning/utilities/device_dtype_mixin.py
+++ b/pytorch_lightning/utilities/device_dtype_mixin.py
@@ -25,7 +25,7 @@ class DeviceDtypeModuleMixin(Module):
 
     def __init__(self) -> None:
         super().__init__()
-        self._dtype = torch.get_default_dtype()
+        self._dtype: Union[str, torch.dtype] = torch.get_default_dtype()
         self._device = torch.device('cpu')
 
     @property
@@ -188,7 +188,7 @@ def __update_properties(
         self, device: Optional[torch.device] = None, dtype: Optional[Union[str, torch.dtype]] = None
     ) -> None:
 
-        def apply_fn(module):
+        def apply_fn(module: Module) -> None:
             if not isinstance(module, DeviceDtypeModuleMixin):
                 return
             if device is not None:
diff --git a/pytorch_lightning/utilities/distributed.py b/pytorch_lightning/utilities/distributed.py
index 28723bbbfa291..cd69ab6732c5e 100644
--- a/pytorch_lightning/utilities/distributed.py
+++ b/pytorch_lightning/utilities/distributed.py
@@ -16,7 +16,7 @@
 import os
 import warnings
 from functools import partial, wraps
-from typing import Any, Callable, List, Optional, Tuple, Union
+from typing import Any, Callable, List, Optional, Tuple, Type, Union
 
 import torch
 from torch.nn.parallel.distributed import DistributedDataParallel
@@ -175,7 +175,11 @@ def sync_ddp(
 class AllGatherGrad(torch.autograd.Function):
 
     @staticmethod
-    def forward(ctx: torch.autograd.Function, tensor: torch.Tensor, group=group.WORLD) -> torch.Tensor:
+    def forward(
+        ctx: torch.autograd.Function,
+        tensor: torch.Tensor,
+        group: Optional['torch.distributed.ProcessGroup'] = group.WORLD,
+    ) -> torch.Tensor:
         ctx.group = group
 
         gathered_tensor = [torch.zeros_like(tensor) for _ in range(torch.distributed.get_world_size())]
@@ -197,7 +201,7 @@ def backward(ctx: torch.autograd.Function, *grad_output: torch.Tensor) -> Tuple[
 
 
 def all_gather_ddp_if_available(
-    tensor: torch.Tensor, group: Optional[Any] = None, sync_grads: bool = False
+    tensor: torch.Tensor, group: Optional[Type['torch.distributed.ProcessGroup']] = None, sync_grads: bool = False
 ) -> torch.Tensor:
     """
     Function to gather a tensor from several distributed processes
diff --git a/pytorch_lightning/utilities/memory.py b/pytorch_lightning/utilities/memory.py
index 1a0dbcd04dff3..7bbac2d443618 100644
--- a/pytorch_lightning/utilities/memory.py
+++ b/pytorch_lightning/utilities/memory.py
@@ -17,10 +17,12 @@
 
 import torch
 
-RECURSIVE_DICT_WITH_TENSORS = Dict[Any, Union[torch.Tensor, 'RECURSIVE_DICT_WITH_TENSORS']]
+RECURSIVE_DICT_WITH_TENSORS = Union[Dict[str, torch.Tensor], Dict[Any, Any]]
 
 
-def recursive_detach(in_dict: RECURSIVE_DICT_WITH_TENSORS, to_cpu: bool = False) -> Dict[str, torch.Tensor]:
+def recursive_detach(
+    in_dict: RECURSIVE_DICT_WITH_TENSORS, to_cpu: bool = False
+) -> Dict[str, Union[Any, Dict[str, torch.Tensor], torch.Tensor]]:
     """Detach all tensors in `in_dict`.
 
     May operate recursively if some of the values in `in_dict` are dictionaries
diff --git a/pytorch_lightning/utilities/metrics.py b/pytorch_lightning/utilities/metrics.py
index 1fd97eea0fc85..aa4a60e08bde4 100644
--- a/pytorch_lightning/utilities/metrics.py
+++ b/pytorch_lightning/utilities/metrics.py
@@ -13,16 +13,16 @@
 # limitations under the License.
 """Helper functions to operate on metric values. """
 
-from typing import Dict, Union
+from typing import Any, Dict, Union
 
 import torch
 
 from pytorch_lightning.utilities.exceptions import MisconfigurationException
 
-RECURSIVE_DICT_WITH_TENSORS = Dict[str, Union[torch.Tensor, 'RECURSIVE_DICT_WITH_TENSORS']]
+RECURSIVE_DICT_WITH_TENSORS = Union[Dict[str, torch.Tensor], Dict[Any, Any]]
 
 
-def metrics_to_scalars(metrics: RECURSIVE_DICT_WITH_TENSORS) -> Dict[str, float]:
+def metrics_to_scalars(metrics: RECURSIVE_DICT_WITH_TENSORS) -> Dict[str, Union[Any, Dict[str, float], float]]:
     """ Recursively walk through a dictionary of metrics and convert single-item tensors to scalar values. """
 
     # TODO: this is duplicated in MetricsHolder. should be unified
diff --git a/pytorch_lightning/utilities/parsing.py b/pytorch_lightning/utilities/parsing.py
index fc7b24a6b286b..38d082d378932 100644
--- a/pytorch_lightning/utilities/parsing.py
+++ b/pytorch_lightning/utilities/parsing.py
@@ -17,7 +17,7 @@
 import types
 from argparse import Namespace
 from types import FrameType
-from typing import Any, Dict, List, Optional, Sequence, Tuple, Union
+from typing import Any, Dict, List, Optional, Sequence, Tuple, Type, Union
 
 from typing_extensions import Literal
 
@@ -105,7 +105,7 @@ def clean_namespace(hparams: Union[Dict[str, Any], Namespace]) -> None:
         del hparams_dict[k]
 
 
-def parse_class_init_keys(cls) -> Tuple[str, str, str]:
+def parse_class_init_keys(cls: Type['pl.LightningModule']) -> Tuple[str, str, str]:
     """Parse key words for standard self, *args and **kwargs
 
     >>> class Model():
diff --git a/pytorch_lightning/utilities/warnings.py b/pytorch_lightning/utilities/warnings.py
index 6dfb67133cf19..08b4d9d962f43 100644
--- a/pytorch_lightning/utilities/warnings.py
+++ b/pytorch_lightning/utilities/warnings.py
@@ -11,7 +11,7 @@
 # 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 typing import Any
+from typing import Any, Set
 
 from pytorch_lightning.utilities.distributed import rank_zero_warn
 
@@ -19,7 +19,7 @@
 class WarningCache:
 
     def __init__(self) -> None:
-        self.warnings = set()
+        self.warnings: Set[Any] = set()
 
     def warn(self, m: Any, *args: Any, **kwargs: Any) -> None:
         if m not in self.warnings:

From 3117438449c792e90a5641e7aafe366bf197059d Mon Sep 17 00:00:00 2001
From: "pre-commit-ci[bot]"
 <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Date: Sat, 15 May 2021 18:38:11 +0000
Subject: [PATCH 45/51] [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci
---
 pytorch_lightning/utilities/distributed.py | 4 +++-
 pytorch_lightning/utilities/memory.py      | 5 ++---
 2 files changed, 5 insertions(+), 4 deletions(-)

diff --git a/pytorch_lightning/utilities/distributed.py b/pytorch_lightning/utilities/distributed.py
index cd69ab6732c5e..8285759eaf473 100644
--- a/pytorch_lightning/utilities/distributed.py
+++ b/pytorch_lightning/utilities/distributed.py
@@ -201,7 +201,9 @@ def backward(ctx: torch.autograd.Function, *grad_output: torch.Tensor) -> Tuple[
 
 
 def all_gather_ddp_if_available(
-    tensor: torch.Tensor, group: Optional[Type['torch.distributed.ProcessGroup']] = None, sync_grads: bool = False
+    tensor: torch.Tensor,
+    group: Optional[Type['torch.distributed.ProcessGroup']] = None,
+    sync_grads: bool = False
 ) -> torch.Tensor:
     """
     Function to gather a tensor from several distributed processes
diff --git a/pytorch_lightning/utilities/memory.py b/pytorch_lightning/utilities/memory.py
index 7bbac2d443618..bfed10fe871ee 100644
--- a/pytorch_lightning/utilities/memory.py
+++ b/pytorch_lightning/utilities/memory.py
@@ -20,9 +20,8 @@
 RECURSIVE_DICT_WITH_TENSORS = Union[Dict[str, torch.Tensor], Dict[Any, Any]]
 
 
-def recursive_detach(
-    in_dict: RECURSIVE_DICT_WITH_TENSORS, to_cpu: bool = False
-) -> Dict[str, Union[Any, Dict[str, torch.Tensor], torch.Tensor]]:
+def recursive_detach(in_dict: RECURSIVE_DICT_WITH_TENSORS,
+                     to_cpu: bool = False) -> Dict[str, Union[Any, Dict[str, torch.Tensor], torch.Tensor]]:
     """Detach all tensors in `in_dict`.
 
     May operate recursively if some of the values in `in_dict` are dictionaries

From 1888663f35a0d3a72b714b8500edcbbdc65184c5 Mon Sep 17 00:00:00 2001
From: tchaton <thomas@grid.ai>
Date: Mon, 17 May 2021 09:10:17 +0100
Subject: [PATCH 46/51] typo

---
 docs/source/governance.rst | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/docs/source/governance.rst b/docs/source/governance.rst
index 1f59491781468..60476822cd59c 100644
--- a/docs/source/governance.rst
+++ b/docs/source/governance.rst
@@ -37,7 +37,4 @@ Alumni
 - Jeff Ling (`jeffling <https://github.com/jeffling>`_)
 - Teddy Koker (`teddykoker <https://github.com/teddykoker>`_)
 - Nate Raw (`nateraw <https://github.com/nateraw>`_)
-<<<<<<< HEAD
-=======
 - Peter Yu (`yukw777 <https://github.com/yukw777>`_)
->>>>>>> upstream/master

From 4ae0c144700f429a0c8bf817c1c27e195311ff63 Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Tue, 18 May 2021 17:18:19 +0200
Subject: [PATCH 47/51] Fix pl.LightningModule type in parsing.py

---
 pytorch_lightning/utilities/parsing.py | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/pytorch_lightning/utilities/parsing.py b/pytorch_lightning/utilities/parsing.py
index 38d082d378932..886a5078c60bf 100644
--- a/pytorch_lightning/utilities/parsing.py
+++ b/pytorch_lightning/utilities/parsing.py
@@ -199,7 +199,7 @@ def save_hyperparameters(
     ignore: Optional[Union[Sequence[str], str]] = None,
     frame: Optional[types.FrameType] = None,
 ) -> None:
-    """See :meth:`~pytorch_lightning.'pl.core.lightning.LightningModule'.save_hyperparameters`"""
+    """See :meth:`~pytorch_lightning.core.lightning.LightningModule'.save_hyperparameters`"""
 
     if len(args) == 1 and not isinstance(args, str) and not args[0]:
         # args[0] is an empty container
@@ -274,14 +274,14 @@ def __repr__(self) -> str:
         return out
 
 
-def _lightning_get_all_attr_holders(model: 'pl.core.lightning.LightningModule', attribute: str) -> Any:
+def _lightning_get_all_attr_holders(model: 'pl.LightningModule', attribute: str) -> Any:
     """
     Special attribute finding for Lightning. Gets all of the objects or dicts that holds attribute.
     Checks for attribute in model namespace, the old hparams namespace/dict, and the datamodule.
     """
     trainer = getattr(model, 'trainer', None)
 
-    holders: List[Union[Dict[Any, Any], Namespace, 'pl.core.lightning.LightningModule']] = []
+    holders: List[Union[Dict[Any, Any], Namespace, 'pl.LightningModule']] = []
 
     # Check if attribute in model
     if hasattr(model, attribute):
@@ -299,7 +299,7 @@ def _lightning_get_all_attr_holders(model: 'pl.core.lightning.LightningModule',
     return holders
 
 
-def _lightning_get_first_attr_holder(model: 'pl.core.lightning.LightningModule', attribute: str) -> Optional[Any]:
+def _lightning_get_first_attr_holder(model: 'pl.LightningModule', attribute: str) -> Optional[Any]:
     """
     Special attribute finding for Lightning.  Gets the object or dict that holds attribute, or None.
     Checks for attribute in model namespace, the old hparams namespace/dict, and the datamodule,
@@ -312,7 +312,7 @@ def _lightning_get_first_attr_holder(model: 'pl.core.lightning.LightningModule',
     return holders[-1]
 
 
-def lightning_hasattr(model: 'pl.core.lightning.LightningModule', attribute: str) -> bool:
+def lightning_hasattr(model: 'pl.LightningModule', attribute: str) -> bool:
     """
     Special hasattr for Lightning. Checks for attribute in model namespace,
     the old hparams namespace/dict, and the datamodule.
@@ -320,7 +320,7 @@ def lightning_hasattr(model: 'pl.core.lightning.LightningModule', attribute: str
     return _lightning_get_first_attr_holder(model, attribute) is not None
 
 
-def lightning_getattr(model: 'pl.core.lightning.LightningModule', attribute: str) -> Any:
+def lightning_getattr(model: 'pl.LightningModule', attribute: str) -> Any:
     """
     Special getattr for Lightning. Checks for attribute in model namespace,
     the old hparams namespace/dict, and the datamodule.
@@ -342,7 +342,7 @@ def lightning_getattr(model: 'pl.core.lightning.LightningModule', attribute: str
     return getattr(holder, attribute)
 
 
-def lightning_setattr(model: 'pl.core.lightning.LightningModule', attribute: str, value: Any) -> None:
+def lightning_setattr(model: 'pl.LightningModule', attribute: str, value: Any) -> None:
     """
     Special setattr for Lightning. Checks for attribute in model namespace
     and the old hparams namespace/dict.

From 9a71d85dcd6370d9253f197a2e3738b0085e28cf Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Fri, 21 May 2021 09:03:02 +0200
Subject: [PATCH 48/51] Add back deleted MisconfigurationException (device)

---
 pytorch_lightning/utilities/apply_func.py | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

diff --git a/pytorch_lightning/utilities/apply_func.py b/pytorch_lightning/utilities/apply_func.py
index 29d0bdf935466..2d24776a41827 100644
--- a/pytorch_lightning/utilities/apply_func.py
+++ b/pytorch_lightning/utilities/apply_func.py
@@ -21,6 +21,7 @@
 import numpy as np
 import torch
 
+from pytorch_lightning.utilities.exceptions import MisconfigurationException
 from pytorch_lightning.utilities.imports import _compare_version, _TORCHTEXT_AVAILABLE
 
 if _TORCHTEXT_AVAILABLE:
@@ -35,12 +36,16 @@
 def to_dtype_tensor(
     value: Union[int, float, List[Union[int, float]]],
     dtype: Optional[torch.dtype] = None,
-    device: Optional[Union[str, torch.device]] = None,
+    device: Union[str, torch.device] = None,
 ) -> torch.Tensor:
+    if device is None:
+        raise MisconfigurationException("device (torch.device) should be provided.")
     return torch.tensor(value, dtype=dtype, device=device)
 
 
-def from_numpy(value: np.ndarray, device: Optional[Union[str, torch.device]] = None) -> torch.Tensor:
+def from_numpy(value: np.ndarray, device: Union[str, torch.device] = None) -> torch.Tensor:
+    if device is None:
+        raise MisconfigurationException("device (torch.device) should be provided.")
     return torch.from_numpy(value).to(device)
 
 
@@ -167,7 +172,9 @@ def batch_to(data: Any) -> Any:
     return apply_to_collection(batch, dtype=dtype, function=batch_to)
 
 
-def convert_to_tensors(data: Any, device: Optional[Union[str, torch.device]] = None) -> Any:
+def convert_to_tensors(data: Any, device: Union[str, torch.device] = None) -> Any:
+    if device is None:
+        raise MisconfigurationException("device (torch.device) should be provided.")
     for src_dtype, conversion_func in CONVERSION_DTYPES:
         data = apply_to_collection(data, src_dtype, partial(conversion_func, device=device))
 

From b7f9ca703f35a00d92cca1b77e9f8890ae16b359 Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Tue, 25 May 2021 15:13:59 +0200
Subject: [PATCH 49/51] [WIP] Tackle some other mypy issues

---
 pytorch_lightning/utilities/apply_func.py         |  2 +-
 pytorch_lightning/utilities/device_dtype_mixin.py |  2 +-
 pytorch_lightning/utilities/distributed.py        |  4 ++--
 pytorch_lightning/utilities/seed.py               | 14 +++++++-------
 4 files changed, 11 insertions(+), 11 deletions(-)

diff --git a/pytorch_lightning/utilities/apply_func.py b/pytorch_lightning/utilities/apply_func.py
index 2d24776a41827..2df8c7cad83e9 100644
--- a/pytorch_lightning/utilities/apply_func.py
+++ b/pytorch_lightning/utilities/apply_func.py
@@ -63,7 +63,7 @@ def apply_to_collection(
     dtype: Union[type, Tuple[type]],
     function: Callable,
     *args: Any,
-    wrong_dtype: Optional[Union[torch.dtype, Tuple[torch.dtype]]] = None,
+    wrong_dtype: Optional[Union[type, Tuple[type]]] = None,
     **kwargs: Any,
 ) -> Any:
     """
diff --git a/pytorch_lightning/utilities/device_dtype_mixin.py b/pytorch_lightning/utilities/device_dtype_mixin.py
index 52c78d2728f84..06c18a7bfedc9 100644
--- a/pytorch_lightning/utilities/device_dtype_mixin.py
+++ b/pytorch_lightning/utilities/device_dtype_mixin.py
@@ -45,7 +45,7 @@ def device(self) -> Union[str, torch.device]:
 
         return device
 
-    def to(self, *args: Any, **kwargs: Any) -> Module:
+    def to(self, *args: Any, **kwargs: Any) -> 'DeviceDtypeModuleMixin':
         """Moves and/or casts the parameters and buffers.
 
         This can be called as
diff --git a/pytorch_lightning/utilities/distributed.py b/pytorch_lightning/utilities/distributed.py
index 8285759eaf473..9997a68515a71 100644
--- a/pytorch_lightning/utilities/distributed.py
+++ b/pytorch_lightning/utilities/distributed.py
@@ -31,10 +31,10 @@
 
 else:
 
-    class ReduceOp:
+    class ReduceOp:  # type: ignore # (see https://github.com/python/mypy/issues/1153)
         SUM = None
 
-    class group:
+    class group:  # type: ignore
         WORLD = None
 
 
diff --git a/pytorch_lightning/utilities/seed.py b/pytorch_lightning/utilities/seed.py
index 78abd06e3b289..05c307769bb7f 100644
--- a/pytorch_lightning/utilities/seed.py
+++ b/pytorch_lightning/utilities/seed.py
@@ -48,13 +48,13 @@ def seed_everything(seed: Optional[int] = None, workers: bool = False) -> int:
     max_seed_value = np.iinfo(np.uint32).max
     min_seed_value = np.iinfo(np.uint32).min
 
-    try:
-        if seed is None:
-            seed = os.environ.get("PL_GLOBAL_SEED")
-        seed = int(seed)
-    except (TypeError, ValueError):
-        seed = _select_seed_randomly(min_seed_value, max_seed_value)
-        rank_zero_warn(f"No correct seed found, seed set to {seed}")
+    if seed is None:
+        global_seed = os.environ.get("PL_GLOBAL_SEED")
+        if isinstance(global_seed, str):
+            seed = int(global_seed)
+        else:
+            seed = _select_seed_randomly(min_seed_value, max_seed_value)
+            rank_zero_warn(f"No correct seed found, seed set to {seed}")
 
     if not (min_seed_value <= seed <= max_seed_value):
         rank_zero_warn(f"{seed} is not in bounds, numpy accepts from {min_seed_value} to {max_seed_value}")

From d2e06b3c5cefe8813ba0c55bd2c15c51f48023c3 Mon Sep 17 00:00:00 2001
From: "Stancl, Daniel" <daniel.stancl@firma.seznam.cz>
Date: Thu, 27 May 2021 23:19:30 +0200
Subject: [PATCH 50/51] Fix some issues after a review

---
 pytorch_lightning/utilities/device_parser.py |  6 +++---
 pytorch_lightning/utilities/memory.py        | 10 +++++-----
 pytorch_lightning/utilities/parsing.py       |  1 +
 3 files changed, 9 insertions(+), 8 deletions(-)

diff --git a/pytorch_lightning/utilities/device_parser.py b/pytorch_lightning/utilities/device_parser.py
index 7111e1be7c288..93d7d0731f6a9 100644
--- a/pytorch_lightning/utilities/device_parser.py
+++ b/pytorch_lightning/utilities/device_parser.py
@@ -12,7 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 import operator
-from typing import Any, Iterable, List, MutableSequence, Optional, Set, Tuple, Union
+from typing import Any, Iterable, List, MutableSequence, Optional, Sequence, Set, Tuple, Union
 
 import torch
 
@@ -194,11 +194,11 @@ def _tpu_cores_valid(tpu_cores: Optional[Union[int, Iterable[int], List[int], Tu
 
     # First condition is necessary for mypy compatibility;
     # list_tpu_cores is required to declare for mypy compatiblity too
-    if (tpu_cores is not None) and (not isinstance(tpu_cores, int)):
+    if tpu_cores is not None and not isinstance(tpu_cores, int):
         list_tpu_cores: List[int] = list(tpu_cores)
 
     # allow picking 1 of 8 indexes
-    if isinstance(list_tpu_cores, List):
+    if isinstance(list_tpu_cores, Sequence):
         has_1_tpu_idx = len(list_tpu_cores) == 1
         is_valid_tpu_idx = list_tpu_cores[0] in range(1, 9)
 
diff --git a/pytorch_lightning/utilities/memory.py b/pytorch_lightning/utilities/memory.py
index bfed10fe871ee..9f0ed0f9b9c3c 100644
--- a/pytorch_lightning/utilities/memory.py
+++ b/pytorch_lightning/utilities/memory.py
@@ -21,7 +21,7 @@
 
 
 def recursive_detach(in_dict: RECURSIVE_DICT_WITH_TENSORS,
-                     to_cpu: bool = False) -> Dict[str, Union[Any, Dict[str, torch.Tensor], torch.Tensor]]:
+                     to_cpu: bool = False,) -> Dict[str, Union[Any, Dict[str, torch.Tensor], torch.Tensor]]:
     """Detach all tensors in `in_dict`.
 
     May operate recursively if some of the values in `in_dict` are dictionaries
@@ -47,14 +47,14 @@ def recursive_detach(in_dict: RECURSIVE_DICT_WITH_TENSORS,
     return out_dict
 
 
-def is_oom_error(exception: Any) -> bool:
+def is_oom_error(exception: Exception) -> bool:
     return is_cuda_out_of_memory(exception) \
         or is_cudnn_snafu(exception) \
         or is_out_of_cpu_memory(exception)
 
 
 # based on https://github.com/BlackHC/toma/blob/master/toma/torch_cuda_memory.py
-def is_cuda_out_of_memory(exception: Any) -> bool:
+def is_cuda_out_of_memory(exception: Exception) -> bool:
     return isinstance(exception, RuntimeError) \
         and len(exception.args) == 1 \
         and "CUDA" in exception.args[0] \
@@ -62,7 +62,7 @@ def is_cuda_out_of_memory(exception: Any) -> bool:
 
 
 # based on https://github.com/BlackHC/toma/blob/master/toma/torch_cuda_memory.py
-def is_cudnn_snafu(exception: Any) -> bool:
+def is_cudnn_snafu(exception: Exception) -> bool:
     # For/because of https://github.com/pytorch/pytorch/issues/4107
     return isinstance(exception, RuntimeError) \
         and len(exception.args) == 1 \
@@ -70,7 +70,7 @@ def is_cudnn_snafu(exception: Any) -> bool:
 
 
 # based on https://github.com/BlackHC/toma/blob/master/toma/cpu_memory.py
-def is_out_of_cpu_memory(exception: Any) -> bool:
+def is_out_of_cpu_memory(exception: Exception) -> bool:
     return isinstance(exception, RuntimeError) \
         and len(exception.args) == 1 \
         and "DefaultCPUAllocator: can't allocate memory" in exception.args[0]
diff --git a/pytorch_lightning/utilities/parsing.py b/pytorch_lightning/utilities/parsing.py
index 886a5078c60bf..18df53bbb4531 100644
--- a/pytorch_lightning/utilities/parsing.py
+++ b/pytorch_lightning/utilities/parsing.py
@@ -166,6 +166,7 @@ def collect_init_args(frame: FrameType, path_args: List[Dict[str, Any]], inside:
           most specific class in the hierarchy.
     """
     _, _, _, local_vars = inspect.getargvalues(frame)
+    # frame.f_back must be of a type FrameType for get_init_args/collect_init_args due to mypy
     if isinstance(frame.f_back, FrameType):
         if '__class__' in local_vars:
             local_args = get_init_args(frame)

From eb902c7ce0da9e3a1c1f5fb841f04e581e81b3da Mon Sep 17 00:00:00 2001
From: "pre-commit-ci[bot]"
 <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Date: Thu, 27 May 2021 21:20:31 +0000
Subject: [PATCH 51/51] [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci
---
 pytorch_lightning/utilities/memory.py | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/pytorch_lightning/utilities/memory.py b/pytorch_lightning/utilities/memory.py
index 9f0ed0f9b9c3c..65858fe8a6d1d 100644
--- a/pytorch_lightning/utilities/memory.py
+++ b/pytorch_lightning/utilities/memory.py
@@ -20,8 +20,10 @@
 RECURSIVE_DICT_WITH_TENSORS = Union[Dict[str, torch.Tensor], Dict[Any, Any]]
 
 
-def recursive_detach(in_dict: RECURSIVE_DICT_WITH_TENSORS,
-                     to_cpu: bool = False,) -> Dict[str, Union[Any, Dict[str, torch.Tensor], torch.Tensor]]:
+def recursive_detach(
+    in_dict: RECURSIVE_DICT_WITH_TENSORS,
+    to_cpu: bool = False,
+) -> Dict[str, Union[Any, Dict[str, torch.Tensor], torch.Tensor]]:
     """Detach all tensors in `in_dict`.
 
     May operate recursively if some of the values in `in_dict` are dictionaries