Restore trainer.current_epoch after tuning (#7434)

* Add a test

* Save and restore current_epoch

* Update CHANGELOG

* alphabetical order

(cherry picked from commit 710b144)

Author: akihironitta

CHANGELOG.md

@@ -25,6 +25,9 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 - Fixed DeepSpeed with IterableDatasets ([#7362](https://github.com/PyTorchLightning/pytorch-lightning/pull/7362))
 
 
+- Fixed `Trainer.current_epoch` not getting restored after tuning ([#7434](https://github.com/PyTorchLightning/pytorch-lightning/pull/7434))
+
+
 ## [1.3.0] - 2021-05-06
 
 ### Added

pytorch_lightning/tuner/lr_finder.py

@@ -288,6 +288,7 @@ def __lr_finder_dump_params(trainer, model):
         'logger': trainer.logger,
         'max_steps': trainer.max_steps,
         'checkpoint_callback': trainer.checkpoint_callback,
+        'current_epoch': trainer.current_epoch,
         'configure_optimizers': model.configure_optimizers,
     }
 
@@ -297,6 +298,7 @@ def __lr_finder_restore_params(trainer, model):
     trainer.logger = trainer.__dumped_params['logger']
     trainer.callbacks = trainer.__dumped_params['callbacks']
     trainer.max_steps = trainer.__dumped_params['max_steps']
+    trainer.current_epoch = trainer.__dumped_params['current_epoch']
     model.configure_optimizers = trainer.__dumped_params['configure_optimizers']
     del trainer.__dumped_params
 

tests/tuner/test_lr_finder.py

@@ -77,7 +77,13 @@ def test_trainer_reset_correctly(tmpdir):
     )
 
     changed_attributes = [
-        'callbacks', 'logger', 'max_steps', 'auto_lr_find', 'accumulate_grad_batches', 'checkpoint_callback'
+        'accumulate_grad_batches',
+        'auto_lr_find',
+        'callbacks',
+        'checkpoint_callback',
+        'current_epoch',
+        'logger',
+        'max_steps',
     ]
     expected = {ca: getattr(trainer, ca) for ca in changed_attributes}
     trainer.tuner.lr_find(model, num_training=5)

tests/tuner/test_scale_batch_size.py

@@ -111,13 +111,13 @@ def test_trainer_reset_correctly(tmpdir):
     )
 
     changed_attributes = [
-        'max_steps',
-        'weights_summary',
-        'logger',
         'callbacks',
         'checkpoint_callback',
-        'limit_train_batches',
         'current_epoch',
+        'limit_train_batches',
+        'logger',
+        'max_steps',
+        'weights_summary',
     ]
     expected = {ca: getattr(trainer, ca) for ca in changed_attributes}
     trainer.tuner.scale_batch_size(model, max_trials=5)