Merge branch 'config_val' of github.com:daniellepintz/pytorch-lightni…

…ng into config_val

pytorch_lightning/trainer/configuration_validator.py

@@ -151,9 +151,13 @@ def __verify_eval_loop_configuration(model: "pl.LightningModule", stage: str) ->
     has_step = is_overridden(step_name, model)
 
     if has_loader and not has_step:
-        raise MisconfigurationException(f"You defined `{loader_name}` but have not defined `{step_name}` on your Lightning Module.")
+        raise MisconfigurationException(
+            f"You defined `{loader_name}` but have not defined `{step_name}` on your Lightning Module."
+        )
     if has_step and not has_loader:
-        raise MisconfigurationException(f"You defined `{step_name}` but have not defined `{loader_name}` on your Lightning Module.")
+        raise MisconfigurationException(
+            f"You defined `{step_name}` but have not defined `{loader_name}` on your Lightning Module."
+        )
 
     # ----------------------------------------------
     # verify model does not have

tests/trainer/test_config_validator.py

@@ -52,13 +52,19 @@ def test_fit_val_loop_config(tmpdir):
     trainer = Trainer(default_root_dir=tmpdir, max_epochs=1)
 
     # has val data but not val loop
-    with pytest.raises(MisconfigurationException, match=r"You defined `val_dataloader` but have not defined `validation_step` on your Lightning Module."):
+    with pytest.raises(
+        MisconfigurationException,
+        match=r"You defined `val_dataloader` but have not defined `validation_step` on your Lightning Module.",
+    ):
         model = BoringModel()
         model.validation_step = None
         trainer.fit(model)
 
     # has val loop but no val data
-    with pytest.raises(MisconfigurationException, match=r"You defined `validation_step` but have not defined `val_dataloader` on your Lightning Module."):
+    with pytest.raises(
+        MisconfigurationException,
+        match=r"You defined `validation_step` but have not defined `val_dataloader` on your Lightning Module.",
+    ):
         model = BoringModel()
         model.val_dataloader = None
         trainer.fit(model)
@@ -69,13 +75,19 @@ def test_test_loop_config(tmpdir):
     trainer = Trainer(default_root_dir=tmpdir, max_epochs=1)
 
     # has test loop but no test data
-    with pytest.raises(MisconfigurationException, match=r"You defined `test_step` but have not defined `test_dataloader` on your Lightning Module"):
+    with pytest.raises(
+        MisconfigurationException,
+        match=r"You defined `test_step` but have not defined `test_dataloader` on your Lightning Module",
+    ):
         model = BoringModel()
         model.test_dataloader = None
         trainer.test(model)
 
     # has test data but no test loop
-    with pytest.raises(MisconfigurationException, match=r"You defined `test_dataloader` but have not defined `test_step` on your Lightning Module"):
+    with pytest.raises(
+        MisconfigurationException,
+        match=r"You defined `test_dataloader` but have not defined `test_step` on your Lightning Module",
+    ):
         model = BoringModel()
         model.test_step = None
         trainer.test(model)
@@ -86,13 +98,19 @@ def test_val_loop_config(tmpdir):
     trainer = Trainer(default_root_dir=tmpdir, max_epochs=1)
 
     # has val loop but no val data
-    with pytest.raises(MisconfigurationException, match=r"You defined `validation_step` but have not defined `val_dataloader` on your Lightning Module"):
+    with pytest.raises(
+        MisconfigurationException,
+        match=r"You defined `validation_step` but have not defined `val_dataloader` on your Lightning Module",
+    ):
         model = BoringModel()
         model.val_dataloader = None
         trainer.validate(model)
 
     # has val data but no val loop
-    with pytest.raises(MisconfigurationException, match=r"You defined `val_dataloader` but have not defined `validation_step` on your Lightning Module"):
+    with pytest.raises(
+        MisconfigurationException,
+        match=r"You defined `val_dataloader` but have not defined `validation_step` on your Lightning Module",
+    ):
         model = BoringModel()
         model.validation_step = None
         trainer.validate(model)