~30X slow down with tf-agents 0.19.0

[Dinple]

Hi,

I have run circuit_training using an old build from 2022 using the default setting on Ariane.

After updating to the latest build and updating tf-agents 0.19.0, the training drastically slows down with the default setting on Ariane.

I can see the buffer being filled by the collector jobs. However, the code will always get stuck on learner.run().

This function seemingly comes from a wrapper class for tf_agents.train.learner.

Is this an expected slowdown by using tf-agents 0.19.0, or is something wrong with my environment?