Deprecate early_stop_callback Trainer argument (part 2) (#3845)

* update tests with EarlyStopping default

* imports

* revert legacy tests

* fix test

* revert

* revert

pytorch_lightning/callbacks/early_stopping.py

@@ -181,7 +181,7 @@ class EarlyStopping(Callback):
         current = logs.get(self.monitor)
 
         # when in dev debugging
-        trainer.dev_debugger.track_early_stopping_history(current)
+        trainer.dev_debugger.track_early_stopping_history(self, current)
 
         if not isinstance(current, torch.Tensor):
             current = torch.tensor(current, device=pl_module.device)

pytorch_lightning/utilities/debugging.py

@@ -154,15 +154,14 @@ class InternalDebugger(object):
         self.pbar_added_metrics.append(metrics)
 
     @enabled_only
-    def track_early_stopping_history(self, current):
-        es = self.trainer.early_stop_callback
+    def track_early_stopping_history(self, callback, current):
         debug_dict = {
             'epoch': self.trainer.current_epoch,
             'global_step': self.trainer.global_step,
             'rank': self.trainer.global_rank,
             'current': current,
-            'best': es.best_score,
-            'patience': es.wait_count
+            'best': callback.best_score,
+            'patience': callback.wait_count
         }
         self.early_stopping_history.append(debug_dict)
 

tests/backends/test_ddp_spawn.py

@@ -3,6 +3,7 @@ import torch
 
 import tests.base.develop_pipelines as tpipes
 import tests.base.develop_utils as tutils
+from pytorch_lightning.callbacks import EarlyStopping
 from tests.base import EvalModelTemplate
 from pytorch_lightning.core import memory
 from pytorch_lightning.trainer import Trainer
@@ -15,7 +16,7 @@ def test_multi_gpu_early_stop_ddp_spawn(tmpdir):
 
     trainer_options = dict(
         default_root_dir=tmpdir,
-        early_stop_callback=True,
+        callbacks=[EarlyStopping()],
         max_epochs=50,
         limit_train_batches=10,
         limit_val_batches=10,

tests/callbacks/test_early_stopping.py

@@ -77,7 +77,7 @@ def test_early_stopping_no_extraneous_invocations(tmpdir):
     expected_count = 4
     trainer = Trainer(
         default_root_dir=tmpdir,
-        early_stop_callback=True,
+        callbacks=[EarlyStopping()],
         val_check_interval=1.0,
         max_epochs=expected_count,
     )

tests/models/test_gpu.py

@@ -8,6 +8,7 @@ from torchtext.data import Batch, Dataset, Example, Field, LabelField
 import tests.base.develop_pipelines as tpipes
 import tests.base.develop_utils as tutils
 from pytorch_lightning import Trainer
+from pytorch_lightning.callbacks import EarlyStopping
 from pytorch_lightning.core import memory
 from pytorch_lightning.utilities import device_parser
 from pytorch_lightning.utilities.exceptions import MisconfigurationException
@@ -25,7 +26,7 @@ def test_multi_gpu_early_stop_dp(tmpdir):
 
     trainer_options = dict(
         default_root_dir=tmpdir,
-        early_stop_callback=True,
+        callbacks=[EarlyStopping()],
         max_epochs=50,
         limit_train_batches=10,
         limit_val_batches=10,

tests/models/test_tpu.py

@@ -6,6 +6,7 @@ from torch.utils.data import DataLoader
 import tests.base.develop_pipelines as tpipes
 from pytorch_lightning import Trainer, seed_everything
 from pytorch_lightning.accelerators import TPUBackend
+from pytorch_lightning.callbacks import EarlyStopping
 from pytorch_lightning.utilities.exceptions import MisconfigurationException
 from tests.base import EvalModelTemplate
 from tests.base.datasets import TrialMNIST
@@ -155,7 +156,7 @@ def test_model_tpu_early_stop(tmpdir):
     """Test if single TPU core training works"""
     model = EvalModelTemplate()
     trainer = Trainer(
-        early_stop_callback=True,
+        callbacks=[EarlyStopping()],
         default_root_dir=tmpdir,
         progress_bar_refresh_rate=0,
         max_epochs=50,
@@ -261,7 +262,7 @@ def test_result_obj_on_tpu(tmpdir):
     trainer_options = dict(
         default_root_dir=tmpdir,
         max_epochs=epochs,
-        early_stop_callback=True,
+        callbacks=[EarlyStopping()],
         row_log_interval=2,
         limit_train_batches=batches,
         weights_summary=None,