Andrei Yarmak

I'm getting the same error when I'm trying to replicate [compute_avg_return(eval_env, random_policy, num_eval_episodes)](https://colab.research.google.com/github/tensorflow/agents/blob/master/docs/tutorials/1_dqn_tutorial.ipynb#scrollTo=9bgU6Q6BZ8Bp&line=1&uniqifier=1) line from the official DQN Tutorial for my custom environment. It looks like the error is caused...

@sguada You're exactly right, looks like it was indeed an issue with my environment. First, the observation specs were defined as (5, 5, 3) instead of (3, 5, 5). Second,...

I'm getting a similar error, but I cannot apply the solution by @sguada above, because I need my action to be a tuple of two integers. My action_spec: ``` self._action_spec...

Thanks a lot - this is super helpful @sguada !!

FYI in case helpful - there is actually a way to use `cache.modify` with parameterized fields today, which simply leverages the following two facts about it (you can check them...