Initialize trainer with None (Lightning-AI#4847)

Co-authored-by: Sean Naren <sean.narenthiran@gmail.com> Co-authored-by: chaton <thomas@grid.ai> Co-authored-by: edenlightning <66261195+edenlightning@users.noreply.github.com>
gianscarpe · Dec 8, 2020 · 79ae66d · 79ae66d
1 parent aeaa6b2
commit 79ae66d
Showing 1 changed file with 1 addition and 1 deletion.
diff --git a/pytorch_lightning/accelerators/ddp_accelerator.py b/pytorch_lightning/accelerators/ddp_accelerator.py
@@ -39,7 +39,7 @@
 
 class DDPAccelerator(Accelerator):
 
-    def __init__(self, trainer, cluster_environment=None, ddp_plugin=None):
+    def __init__(self, trainer=None, cluster_environment=None, ddp_plugin=None):
         """
         Runs training using DDP strategy on a single machine (manually, not via cluster start)