Hi, is there any way to disable ClearML (via some env variable or argument), eg. while debugging or when the on-premise service is not available?

Hi, is there any way to disable ClearML when it’s not configured and pointing to the public demo service?

Hi, it seems ClearML has trouble serializing Keras model config for TF >= 2.5 when the model contains a Lambda layer with functions like K.mean. I’ve created...