[Fix] Fix the mismatch torch version of MlflowLoggerHook(#1680)

* [Fix] mlflow logger error

* [Update] mlflow hook test arguments

* Replace torch.__version__ with mmcv.TORCH_VERSION

* update test code

mmcv/runner/hooks/logger/mlflow.py

 # Copyright (c) OpenMMLab. All rights reserved.
+from mmcv.utils import TORCH_VERSION
 from ...dist_utils import master_only
 from ..hook import HOOKS
 from .base import LoggerHook
@@ -73,4 +74,7 @@ class MlflowLoggerHook(LoggerHook):
     @master_only
     def after_run(self, runner):
         if self.log_model:
-            self.mlflow_pytorch.log_model(runner.model, 'models')
+            self.mlflow_pytorch.log_model(
+                runner.model,
+                'models',
+                pip_requirements=[f'torch=={TORCH_VERSION}'])

tests/test_runner/test_hooks.py

@@ -34,6 +34,7 @@ from mmcv.runner.hooks.lr_updater import (CosineRestartLrUpdaterHook,
                                           FlatCosineAnnealingLrUpdaterHook,
                                           OneCycleLrUpdaterHook,
                                           StepLrUpdaterHook)
+from mmcv.utils import TORCH_VERSION
 
 sys.modules['petrel_client'] = MagicMock()
 sys.modules['petrel_client.client'] = MagicMock()
@@ -1392,7 +1393,9 @@ def test_mlflow_hook(log_model):
         }, step=6)
     if log_model:
         hook.mlflow_pytorch.log_model.assert_called_with(
-            runner.model, 'models')
+            runner.model,
+            'models',
+            pip_requirements=[f'torch=={TORCH_VERSION}'])
     else:
         assert not hook.mlflow_pytorch.log_model.called