fix(wandb): use same logger on multiple training loops (#2055)

* fix(wandb): use same logger on multiple training loops

New training loops reset step to 0 which would previously try to overwrite logs

fix #2015

* docs(changelog.md): add reference to PR 2055

CHANGELOG.md

@@ -66,6 +66,8 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 
 - Fixed `save_weights_only` in ModelCheckpoint ([#1780](https://github.com/PyTorchLightning/pytorch-lightning/pull/1780))
 
+- Allow use of same `WandbLogger` instance for multiple training loops ([#2055](https://github.com/PyTorchLightning/pytorch-lightning/pull/2055))
+
 ## [0.7.6] - 2020-05-16
 
 ### Added

pytorch_lightning/loggers/wandb.py

@@ -128,7 +128,7 @@ def log_hyperparams(self, params: Union[Dict[str, Any], Namespace]) -> None:
 
     @rank_zero_only
     def log_metrics(self, metrics: Dict[str, float], step: Optional[int] = None) -> None:
-        self.experiment.log(metrics, step=step)
+        self.experiment.log({'global_step': step, **metrics} if step is not None else metrics)
 
     @property
     def name(self) -> str:

tests/loggers/test_wandb.py

@@ -13,11 +13,11 @@ def test_wandb_logger(wandb):
     logger = WandbLogger(anonymous=True, offline=True)
 
     logger.log_metrics({'acc': 1.0})
-    wandb.init().log.assert_called_once_with({'acc': 1.0}, step=None)
+    wandb.init().log.assert_called_once_with({'acc': 1.0})
 
     wandb.init().log.reset_mock()
     logger.log_metrics({'acc': 1.0}, step=3)
-    wandb.init().log.assert_called_once_with({'acc': 1.0}, step=3)
+    wandb.init().log.assert_called_once_with({'global_step': 3, 'acc': 1.0})
 
     logger.log_hyperparams({'test': None})
     wandb.init().config.update.assert_called_once_with({'test': None}, allow_val_change=True)