Closed vitkl closed 3 years ago
We can filter this warning. First we should reproduce the issue using one of the test cases in test_pyro.py
and then add the filter here:
Though maybe we can also only filter it if it's a Pyro base module instance.
warning is replicated with test_pyro_bayesian_regression_jit
tests/models/test_pyro.py::test_pyro_bayesian_regression_jit
/Users/mjayasur/venv3.8.5/lib/python3.8/site-packages/pytorch_lightning/utilities/distributed.py:50: UserWarning: LightningModule.configure_optimizers
returned None
, this fit will run with no optimizer
warnings.warn(*args, **kwargs)
This warning is not expected. Would be good if you can catch this: