CI: add flake8

.github/workflows/code-formatting.yml

@@ -25,7 +25,7 @@ jobs:
 
   python-types:
     name: Python static type checking with Pyright
-    runs-on: ubuntu-18.04
+    runs-on: ubuntu-20.04
 
     # Timeout: https://stackoverflow.com/a/59076067/4521646
     timeout-minutes: 15
@@ -71,3 +71,24 @@ jobs:
       - name: Run type checking
         run: |
           $(npm bin)/pyright --project .pyrightconfig.json
+
+  python-pep8:
+    name: Python formatting PEP8
+    runs-on: ubuntu-20.04
+
+    # Timeout: https://stackoverflow.com/a/59076067/4521646
+    timeout-minutes: 10
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v2
+      - uses: actions/setup-python@v2
+        with:
+          python-version: 3.7
+
+      - name: Install dependencies
+        run: |
+          pip install flake8
+
+      - name: Run checking
+        run: |
+          flake8 .

pytorch_lightning/accelerators/ddp2_accelerator.py

@@ -16,6 +16,8 @@
 
 import torch
 import torch.distributed as torch_distrib
+from pytorch_lightning.core.lightning import LightningModule
+
 from pytorch_lightning.utilities.exceptions import MisconfigurationException
 from pytorch_lightning.core.step_result import Result
 from pytorch_lightning.distributed.dist import LightningDistributed
@@ -191,14 +193,14 @@ def ddp_train(self, process_idx, mp_queue, model):
         return results
 
     def configure_ddp(
-        self, model: "LightningModule", device_ids: List[int]
+        self, model: LightningModule, device_ids: List[int]
     ) -> DistributedDataParallel:
         model = LightningDistributedDataParallel(
             model, device_ids=device_ids, find_unused_parameters=True
         )
         return model
 
-    def configure_sync_batchnorm(self, model: "LightningModule") -> "LightningModule":
+    def configure_sync_batchnorm(self, model: LightningModule) -> LightningModule:
         """
         Add global batchnorm for a model spread across multiple GPUs and nodes.
 

pytorch_lightning/accelerators/ddp_accelerator.py

@@ -20,7 +20,7 @@
 from time import sleep
 from typing import Optional
 import numpy as np
-
+from pytorch_lightning.core.lightning import LightningModule
 
 from pytorch_lightning import _logger as log
 from pytorch_lightning.utilities.distributed import find_free_network_port
@@ -284,14 +284,14 @@ def ddp_train(self, process_idx, model):
         return results
 
     def configure_ddp(
-        self, model: "LightningModule", device_ids: List[int]
+        self, model: LightningModule, device_ids: List[int]
     ) -> DistributedDataParallel:
         model = LightningDistributedDataParallel(
             model, device_ids=device_ids, find_unused_parameters=True
         )
         return model
 
-    def configure_sync_batchnorm(self, model: "LightningModule") -> "LightningModule":
+    def configure_sync_batchnorm(self, model: LightningModule) -> LightningModule:
         """
         Add global batchnorm for a model spread across multiple GPUs and nodes.
 

pytorch_lightning/accelerators/ddp_cpu_slurm_accelerator.py

@@ -15,6 +15,7 @@
 import torch
 import torch.distributed as torch_distrib
 import torch.distributed as dist
+from pytorch_lightning.core.lightning import LightningModule
 
 from pytorch_lightning.accelerators.accelerator import Accelerator
 from pytorch_lightning import _logger as log
@@ -177,14 +178,14 @@ def ddp_train(self, process_idx, model):
         return results
 
     def configure_ddp(
-        self, model: "LightningModule", device_ids: List[int]
+        self, model: LightningModule, device_ids: List[int]
     ) -> DistributedDataParallel:
         model = LightningDistributedDataParallel(
             model, device_ids=device_ids, find_unused_parameters=True
         )
         return model
 
-    def configure_sync_batchnorm(self, model: "LightningModule") -> "LightningModule":
+    def configure_sync_batchnorm(self, model: LightningModule) -> LightningModule:
         """
         Add global batchnorm for a model spread across multiple GPUs and nodes.
 

pytorch_lightning/accelerators/ddp_cpu_spawn_accelerator.py

@@ -18,6 +18,7 @@
 import torch.distributed as torch_distrib
 import torch.distributed as dist
 import torch.multiprocessing as mp
+from pytorch_lightning.core.lightning import LightningModule
 
 from pytorch_lightning import _logger as log
 from pytorch_lightning.accelerators.accelerator import Accelerator
@@ -210,14 +211,14 @@ def transfer_distrib_spawn_state_on_fit_end(self, model, mp_queue, results):
             mp_queue.put(results)
 
     def configure_ddp(
-        self, model: "LightningModule", device_ids: List[int]
+        self, model: LightningModule, device_ids: List[int]
     ) -> DistributedDataParallel:
         model = LightningDistributedDataParallel(
             model, device_ids=device_ids, find_unused_parameters=True
         )
         return model
 
-    def configure_sync_batchnorm(self, model: "LightningModule") -> "LightningModule":
+    def configure_sync_batchnorm(self, model: LightningModule) -> LightningModule:
         """
         Add global batchnorm for a model spread across multiple GPUs and nodes.
 

pytorch_lightning/accelerators/ddp_cpu_torchelastic_accelerator.py

@@ -15,6 +15,7 @@
 import torch
 import torch.distributed as torch_distrib
 import torch.distributed as dist
+from pytorch_lightning.core.lightning import LightningModule
 
 from pytorch_lightning.accelerators.accelerator import Accelerator
 from pytorch_lightning import _logger as log
@@ -176,14 +177,14 @@ def ddp_train(self, process_idx, model):
         return results
 
     def configure_ddp(
-        self, model: "LightningModule", device_ids: List[int]
+        self, model: LightningModule, device_ids: List[int]
     ) -> DistributedDataParallel:
         model = LightningDistributedDataParallel(
             model, device_ids=device_ids, find_unused_parameters=True
         )
         return model
 
-    def configure_sync_batchnorm(self, model: "LightningModule") -> "LightningModule":
+    def configure_sync_batchnorm(self, model: LightningModule) -> LightningModule:
         """
         Add global batchnorm for a model spread across multiple GPUs and nodes.
 

pytorch_lightning/accelerators/ddp_slurm_accelerator.py

@@ -15,6 +15,7 @@
 import torch
 import torch.distributed as torch_distrib
 import torch.distributed as dist
+from pytorch_lightning.core.lightning import LightningModule
 
 from pytorch_lightning.accelerators.accelerator import Accelerator
 from pytorch_lightning import _logger as log
@@ -182,14 +183,14 @@ def ddp_train(self, process_idx, model):
         return results
 
     def configure_ddp(
-        self, model: "LightningModule", device_ids: List[int]
+        self, model: LightningModule, device_ids: List[int]
     ) -> DistributedDataParallel:
         model = LightningDistributedDataParallel(
             model, device_ids=device_ids, find_unused_parameters=True
         )
         return model
 
-    def configure_sync_batchnorm(self, model: "LightningModule") -> "LightningModule":
+    def configure_sync_batchnorm(self, model: LightningModule) -> LightningModule:
         """
         Add global batchnorm for a model spread across multiple GPUs and nodes.
 

pytorch_lightning/accelerators/ddp_spawn_accelerator.py

@@ -18,6 +18,7 @@
 import torch.multiprocessing as mp
 import torch.distributed as torch_distrib
 import torch.distributed as dist
+from pytorch_lightning.core.lightning import LightningModule
 
 from pytorch_lightning import _logger as log
 from pytorch_lightning.accelerators.accelerator import Accelerator
@@ -237,14 +238,14 @@ def transfer_distrib_spawn_state_on_fit_end(self, model, mp_queue, results):
             mp_queue.put(last_path)
 
     def configure_ddp(
-        self, model: "LightningModule", device_ids: List[int]
+        self, model: LightningModule, device_ids: List[int]
     ) -> DistributedDataParallel:
         model = LightningDistributedDataParallel(
             model, device_ids=device_ids, find_unused_parameters=True
         )
         return model
 
-    def configure_sync_batchnorm(self, model: "LightningModule") -> "LightningModule":
+    def configure_sync_batchnorm(self, model: LightningModule) -> LightningModule:
         """
         Add global batchnorm for a model spread across multiple GPUs and nodes.
 

pytorch_lightning/accelerators/ddp_torchelastic_accelerator.py

@@ -15,6 +15,7 @@
 import torch
 import torch.distributed as torch_distrib
 import torch.distributed as dist
+from pytorch_lightning.core.lightning import LightningModule
 
 from pytorch_lightning.accelerators.accelerator import Accelerator
 from pytorch_lightning import _logger as log
@@ -179,14 +180,14 @@ def ddp_train(self, process_idx, model):
         return results
 
     def configure_ddp(
-        self, model: "LightningModule", device_ids: List[int]
+        self, model: LightningModule, device_ids: List[int]
     ) -> DistributedDataParallel:
         model = LightningDistributedDataParallel(
             model, device_ids=device_ids, find_unused_parameters=True
         )
         return model
 
-    def configure_sync_batchnorm(self, model: "LightningModule") -> "LightningModule":
+    def configure_sync_batchnorm(self, model: LightningModule) -> LightningModule:
         """
         Add global batchnorm for a model spread across multiple GPUs and nodes.
 

pytorch_lightning/accelerators/horovod_accelerator.py

@@ -152,7 +152,7 @@ def test_step(self, args):
         return output
 
     def backward(self, closure_loss, optimizer, opt_idx, *args, **kwargs):
-        super().backward(closure_loss, optimizer, opt_idx,  *args, **kwargs)
+        super().backward(closure_loss, optimizer, opt_idx, *args, **kwargs)
         optimizer.synchronize()
 
     def on_train_epoch_end(self, outputs):

pytorch_lightning/callbacks/early_stopping.py

@@ -35,8 +35,6 @@
 torch_inf = torch.tensor(np.Inf)
 
 
-
-
 class EarlyStopping(Callback):
     r"""
     Monitor a validation metric and stop training when it stops improving.

pytorch_lightning/core/lightning.py

@@ -233,7 +233,7 @@ def log(
                 m = f'on_step=True cannot be used on {self._current_fx_name} method'
                 raise MisconfigurationException(m)
 
-            if 'epoch_end' in self._current_fx_name and on_epoch == False:
+            if 'epoch_end' in self._current_fx_name and on_epoch is False:
                 m = f'on_epoch cannot be False when called from the {self._current_fx_name} method'
                 raise MisconfigurationException(m)
 

pytorch_lightning/plugins/apex.py

@@ -12,6 +12,8 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 from typing import List, Tuple
+
+from pytorch_lightning.core.lightning import LightningModule
 from torch.optim.optimizer import Optimizer
 from pytorch_lightning.utilities.distributed import rank_zero_warn
 from pytorch_lightning.utilities import AMPType
@@ -65,10 +67,10 @@ def backward(self, closure_loss, optimizer, opt_idx, *args, **kwargs):
     def configure_apex(
         self,
         amp: object,
-        model: "LightningModule",
+        model: LightningModule,
         optimizers: List[Optimizer],
         amp_level: str,
-    ) -> Tuple["LightningModule", List[Optimizer]]:
+    ) -> Tuple[LightningModule, List[Optimizer]]:
         r"""
         Override to init AMP your own way.
         Must return a model and list of optimizers.

tests/backends/test_accelerator_connector.py

@@ -310,9 +310,7 @@ def on_fit_start(self, trainer, pl_module):
 @mock.patch('torch.cuda.device_count', return_value=0)
 def test_custom_accelerator(tmpdir):
     class Accel(Accelerator):
-        def init_ddp_connection(
-        self, global_rank: int, world_size: int, is_slurm_managing_tasks: bool = True
-    ) -> None:
+        def init_ddp_connection(self, global_rank: int, world_size: int, is_slurm_managing_tasks: bool = True) -> None:
             pass
 
     class CB(Callback):

tests/loggers/test_mlflow.py

@@ -121,7 +121,7 @@ def test_mlflow_log_dir(client, mlflow, tmpdir):
 def test_mlflow_logger_dirs_creation(tmpdir):
     """ Test that the logger creates the folders and files in the right place. """
     if not importlib.util.find_spec('mlflow'):
-        pytest.xfail(f"test for explicit file creation requires mlflow dependency to be installed.")
+        pytest.xfail("test for explicit file creation requires mlflow dependency to be installed.")
 
     assert not os.listdir(tmpdir)
     logger = MLFlowLogger('test', save_dir=tmpdir)

tests/metrics/classification/test_accuracy.py

@@ -93,9 +93,9 @@ def test_accuracy_invalid_shape():
         (_multiclass_prob_inputs.preds, _multiclass_prob_inputs.target, _sk_accuracy_multiclass_prob),
         (_multiclass_inputs.preds, _multiclass_inputs.target, _sk_accuracy_multiclass),
         (
-                _multidim_multiclass_prob_inputs.preds,
-                _multidim_multiclass_prob_inputs.target,
-                _sk_accuracy_multidim_multiclass_prob,
+            _multidim_multiclass_prob_inputs.preds,
+            _multidim_multiclass_prob_inputs.target,
+            _sk_accuracy_multidim_multiclass_prob,
         ),
         (_multidim_multiclass_inputs.preds, _multidim_multiclass_inputs.target, _sk_accuracy_multidim_multiclass),
     ],

tests/metrics/classification/test_f_beta.py

@@ -90,18 +90,18 @@ def _sk_fbeta_multidim_multiclass(preds, target, average='micro', beta=1.0):
         (_multiclass_prob_inputs.preds, _multiclass_prob_inputs.target, _sk_fbeta_multiclass_prob, NUM_CLASSES, False),
         (_multiclass_inputs.preds, _multiclass_inputs.target, _sk_fbeta_multiclass, NUM_CLASSES, False),
         (
-                _multidim_multiclass_prob_inputs.preds,
-                _multidim_multiclass_prob_inputs.target,
-                _sk_fbeta_multidim_multiclass_prob,
-                NUM_CLASSES,
-                False,
+            _multidim_multiclass_prob_inputs.preds,
+            _multidim_multiclass_prob_inputs.target,
+            _sk_fbeta_multidim_multiclass_prob,
+            NUM_CLASSES,
+            False,
         ),
         (
-                _multidim_multiclass_inputs.preds,
-                _multidim_multiclass_inputs.target,
-                _sk_fbeta_multidim_multiclass,
-                NUM_CLASSES,
-                False,
+            _multidim_multiclass_inputs.preds,
+            _multidim_multiclass_inputs.target,
+            _sk_fbeta_multidim_multiclass,
+            NUM_CLASSES,
+            False,
         ),
     ],
 )

tests/metrics/classification/test_precision_recall.py

@@ -90,18 +90,18 @@ def _sk_prec_recall_multidim_multiclass(preds, target, sk_fn=precision_score, av
         (_multiclass_prob_inputs.preds, _multiclass_prob_inputs.target, _sk_prec_recall_multiclass_prob, NUM_CLASSES, False),
         (_multiclass_inputs.preds, _multiclass_inputs.target, _sk_prec_recall_multiclass, NUM_CLASSES, False),
         (
-                _multidim_multiclass_prob_inputs.preds,
-                _multidim_multiclass_prob_inputs.target,
-                _sk_prec_recall_multidim_multiclass_prob,
-                NUM_CLASSES,
-                False,
+            _multidim_multiclass_prob_inputs.preds,
+            _multidim_multiclass_prob_inputs.target,
+            _sk_prec_recall_multidim_multiclass_prob,
+            NUM_CLASSES,
+            False,
         ),
         (
-                _multidim_multiclass_inputs.preds,
-                _multidim_multiclass_inputs.target,
-                _sk_prec_recall_multidim_multiclass,
-                NUM_CLASSES,
-                False,
+            _multidim_multiclass_inputs.preds,
+            _multidim_multiclass_inputs.target,
+            _sk_prec_recall_multidim_multiclass,
+            NUM_CLASSES,
+            False,
         ),
     ],
 )

tests/metrics/test_metric.py

@@ -72,7 +72,6 @@ def test_add_state_persistent():
         assert "b" not in a.state_dict()
 
 
-
 def test_reset():
     class A(Dummy):
         pass

tests/models/test_hparams.py

@@ -552,7 +552,7 @@ def test_args(tmpdir):
     trainer.fit(model)
 
     raw_checkpoint_path = _raw_checkpoint_path(trainer)
-    with pytest.raises(TypeError, match="__init__\(\) got an unexpected keyword argument 'test'"):
+    with pytest.raises(TypeError, match=r"__init__\(\) got an unexpected keyword argument 'test'"):
         SubClassVarArgs.load_from_checkpoint(raw_checkpoint_path)