# Observation space conflict with wrap\_deepmind

**URL:** <https://discuss.ray.io/t/observation-space-conflict-with-wrap-deepmind/10017>\
**Category:** Configure Algorithm, Training, Evaluation, Scaling\
**Created:** [April 2, 2023, 10:23pm UTC](https://discuss.ray.io/t/observation-space-conflict-with-wrap-deepmind/10017 "2023-04-02T22:23:20Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![ntumm120](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/ntumm120/32/4164_2.png) [@ntumm120](https://discuss.ray.io/u/ntumm120)\
**Post date:** [April 2, 2023, 10:23pm UTC](https://discuss.ray.io/t/observation-space-conflict-with-wrap-deepmind/10017/1 "2023-04-02T22:23:20Z")

</div>

I am trying to use the wrap\_deepmind function on my Gym environment to generate rollouts with the deepmind-transformed observations. However, after I wrap the environment and call compute\_single\_action, I get a value error saying that the observation which is shaped (84, 84, 4) is outside of bounds. But the bounds correspond to the original observation space (prior to wrapping) which is (210, 160, 3). Does anyone know how to reconcile this?

---

<div class="post-metadata">

**Author:** ![arturn](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/arturn/32/2096_2.png) [@arturn](https://discuss.ray.io/u/arturn)\
**Post date:** [May 8, 2023, 11:26pm UTC](https://discuss.ray.io/t/observation-space-conflict-with-wrap-deepmind/10017/2 "2023-05-08T23:26:47Z")

</div>

You are probably calling `Policy.compute_action_action()` instead of `Algorithm.compute_single_action()`. Policy does not hold a wrapped environment or preprocessor to do this transformation.