flush buffer when saving model (#232)

Co-authored-by: xurui <xurui4@sensetime.com>

mmcv/runner/checkpoint.py

@@ -245,5 +245,6 @@ def save_checkpoint(model, filename, optimizer=None, meta=None):
     }
     if optimizer is not None:
         checkpoint['optimizer'] = optimizer.state_dict()
-
-    torch.save(checkpoint, filename)
+    # immediately flush buffer
+    with open(filename, 'wb') as f:
+        torch.save(checkpoint, f)