Adress comments

environment.yml

@@ -21,6 +21,7 @@ dependencies:
     - pytest-cov
     - pytest-flake8
     - flake8
+    - autopep8
     - check-manifest
     - twine==1.13.0
     - pillow<7.0.0

pytorch_lightning/callbacks/__init__.py

@@ -1,4 +1,4 @@
-from .callback import Callback
+from .base import Callback
 from .early_stopping import EarlyStopping
 from .model_checkpoint import ModelCheckpoint
 from .gradient_accumulation_scheduler import GradientAccumulationScheduler

pytorch_lightning/callbacks/callback.py → pytorch_lightning/callbacks/base.py

@@ -5,10 +5,13 @@
 Callbacks supported by Lightning
 """
 
+import abc
+
+
 _NO_TRAINER_ERROR_MSG = ".set_trainer() should be called after the callback initialization"
 
 
-class Callback(object):
+class Callback(abc.ABC):
     """Abstract base class used to build new callbacks."""
 
     def __init__(self):

pytorch_lightning/callbacks/early_stopping.py

@@ -3,7 +3,7 @@
 
 import numpy as np
 
-from .callback import Callback
+from .base import Callback
 
 
 class EarlyStopping(Callback):
@@ -38,9 +38,8 @@ class EarlyStopping(Callback):
         Trainer(early_stop_callback=early_stopping)
     """
 
-    def __init__(self, monitor='val_loss',
-                 min_delta=0.0, patience=0, verbose=0, mode='auto', strict=True):
-        super(EarlyStopping, self).__init__()
+    def __init__(self, monitor='val_loss', min_delta=0.0, patience=0, verbose=0, mode='auto', strict=True):
+        super().__init__()
 
         self.monitor = monitor
         self.patience = patience
@@ -55,20 +54,13 @@ def __init__(self, monitor='val_loss',
                 log.info(f'EarlyStopping mode {mode} is unknown, fallback to auto mode.')
             mode = 'auto'
 
-        if mode == 'min':
-            self.monitor_op = np.less
-        elif mode == 'max':
-            self.monitor_op = np.greater
-        else:
-            if 'acc' in self.monitor:
-                self.monitor_op = np.greater
-            else:
-                self.monitor_op = np.less
-
-        if self.monitor_op == np.greater:
-            self.min_delta *= 1
-        else:
-            self.min_delta *= -1
+        mode_dict = {
+            'min': np.less,
+            'max': np.greater,
+            'auto': np.greater if 'acc' in self.monitor else np.less
+        }
+        self.monitor_op = mode_dict[mode]
+        self.min_delta *= 1 if self.monitor_op == np.greater else -1
 
         self.on_train_begin()
 
@@ -95,7 +87,6 @@ def on_train_begin(self):
         self.best = np.Inf if self.monitor_op == np.less else -np.Inf
 
     def on_epoch_end(self):
-
         logs = self.trainer.callback_metrics
         stop_training = False
         if not self.check_metrics(logs):

pytorch_lightning/callbacks/gradient_accumulation_scheduler.py

@@ -1,6 +1,6 @@
 import warnings
 
-from .callback import Callback
+from .base import CallbackBase
 
 
 class GradientAccumulationScheduler(Callback):
@@ -25,10 +25,10 @@ class GradientAccumulationScheduler(Callback):
     def __init__(self, scheduling: dict):
         super().__init__()
 
-        if scheduling == {}:  # empty dict error
+        if not scheduling:  # empty dict error
             raise TypeError("Empty dict cannot be interpreted correct")
 
-        for key in scheduling.keys():
+        for key in scheduling:
             if not isinstance(key, int) or not isinstance(scheduling[key], int):
                 raise TypeError("All epoches and accumulation factor must be integers")
 
@@ -45,7 +45,6 @@ def __init__(self, scheduling: dict):
         self.epochs = sorted(scheduling.keys())
 
     def on_epoch_begin(self):
-
         trainer = self.trainer
         # indexing epochs from 1 (until v0.6.x)
         # In v0.8.0, ` + 1` should be removed.

pytorch_lightning/callbacks/model_checkpoint.py

@@ -5,12 +5,11 @@
 
 import numpy as np
 
-from .callback import Callback
+from .base import Callback
 
 
 class ModelCheckpoint(Callback):
     r"""
-
     Save the model after every epoch.
 
     Args:
@@ -27,14 +26,14 @@ class ModelCheckpoint(Callback):
         save_top_k (int): if `save_top_k == k`,
             the best k models according to
             the quantity monitored will be saved.
-            if `save_top_k == 0`, no models are saved.
-            if `save_top_k == -1`, all models are saved.
+            if ``save_top_k == 0``, no models are saved.
+            if ``save_top_k == -1``, all models are saved.
             Please note that the monitors are checked every `period` epochs.
-            if `save_top_k >= 2` and the callback is called multiple
+            if ``save_top_k >= 2`` and the callback is called multiple
             times inside an epoch, the name of the saved file will be
             appended with a version count starting with `v0`.
         mode (str): one of {auto, min, max}.
-            If `save_top_k != 0`, the decision
+            If ``save_top_k != 0``, the decision
             to overwrite the current save file is made
             based on either the maximization or the
             minimization of the monitored quantity. For `val_acc`,
@@ -60,11 +59,11 @@ class ModelCheckpoint(Callback):
     def __init__(self, filepath, monitor='val_loss', verbose=0,
                  save_top_k=1, save_weights_only=False,
                  mode='auto', period=1, prefix=''):
-        super(ModelCheckpoint, self).__init__()
+        super().__init__()
         if (
-            save_top_k and
-            os.path.isdir(filepath) and
-            len(os.listdir(filepath)) > 0
+            save_top_k
+            and os.path.isdir(filepath)
+            and len(os.listdir(filepath)) > 0
         ):
             warnings.warn(
                 f"Checkpoint directory {filepath} exists and is not empty with save_top_k != 0."
@@ -111,34 +110,26 @@ def __init__(self, filepath, monitor='val_loss', verbose=0,
                 self.mode = 'min'
 
     def _del_model(self, filepath):
-        dirpath = os.path.dirname(filepath)
-
-        # make paths
-        os.makedirs(dirpath, exist_ok=True)
-
         try:
             shutil.rmtree(filepath)
         except OSError:
             os.remove(filepath)
 
     def _save_model(self, filepath):
-        dirpath = os.path.dirname(filepath)
-
         # make paths
-        os.makedirs(dirpath, exist_ok=True)
+        os.makedirs(os.path.dirname(filepath), exist_ok=True)
 
         # delegate the saving to the model
         assert self.save_function is not None, ".save_function() not set"
         self.save_function(filepath)
 
     def check_monitor_top_k(self, current):
-        less_than_k_models = len(self.best_k_models.keys()) < self.save_top_k
+        less_than_k_models = len(self.best_k_models) < self.save_top_k
         if less_than_k_models:
             return True
         return self.monitor_op(current, self.best_k_models[self.kth_best_model])
 
     def on_validation_end(self):
-
         logs = self.trainer.callback_metrics
         epoch = self.trainer.current_epoch
         self.epochs_since_last_check += 1
@@ -174,18 +165,13 @@ def on_validation_end(self):
                         self.best_k_models[filepath] = current
                         if len(self.best_k_models.keys()) == self.save_top_k:
                             # monitor dict has reached k elements
-                            if self.mode == 'min':
-                                self.kth_best_model = max(
-                                    self.best_k_models, key=self.best_k_models.get)
-                            else:
-                                self.kth_best_model = min(
-                                    self.best_k_models, key=self.best_k_models.get)
+                            _op = min if self.mode == 'min' else max
+                            self.kth_best_model = _op(self.best_k_models, key=self.best_k_models.get)
                             self.kth_value = self.best_k_models[self.kth_best_model]
 
-                        if self.mode == 'min':
-                            self.best = min(self.best_k_models.values())
-                        else:
-                            self.best = max(self.best_k_models.values())
+                        _op = min if self.mode == 'min' else max
+                        self.best = _op(self.best_k_models.values())
+
                         if self.verbose > 0:
                             log.info(
                                 f'\nEpoch {epoch:05d}: {self.monitor} reached'