What is the problem?

I haven't used it in a while, but it seems to me that the lr_schedule no longer get's applied when running IMPALA, the attached script works as expected with PPO, but the reported lr remains constant for IMPALA, unless it's just a reporting issue.

Ray version and other system information (Python version, TensorFlow version, OS):
ray 0.8.6, python 3.6.8, tf 2.2, os rhel 7.6

Reproduction (REQUIRED)

Please provide a script that can be run to reproduce the issue. The script should have no external library dependencies (i.e., use fake or mock data / environments):

If we cannot run your script, we cannot fix your issue.

• I have verified my script runs in a clean environment and reproduces the issue.
• I have verified the issue also occurs with the latest wheels.

cart-pole-impala.zip