# Save experience from custom policy

**URL:** <https://discuss.ray.io/t/save-experience-from-custom-policy/6751>\
**Category:** RLlib\
**Created:** [July 7, 2022, 5:51am UTC](https://discuss.ray.io/t/save-experience-from-custom-policy/6751 "2022-07-07T05:51:12Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![michele-pel](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/michele-pel/32/2844_2.png) [@michele-pel](https://discuss.ray.io/u/michele-pel)\
**Post date:** [July 7, 2022, 5:51am UTC](https://discuss.ray.io/t/save-experience-from-custom-policy/6751/1 "2022-07-07T05:51:12Z")

</div>

**How severe does this issue affect your experience of using Ray?**

- Medium: It contributes to significant difficulty to complete my task, but I can work around it.

Goodmornig,  
I am creating a custom policy and I want to use it for producing some experience replay.

_Converting external experiences to batch format_ is quite limited and old: I am using a multiagent environment with an rnn so I need to save also the state information and it lacks of a postprocessing. [https://docs.ray.io/en/latest/rllib/rllib-offline](https://docs.ray.io/en/latest/rllib/rllib-offline).

even adapting the script proposed to my case, if I use the batches produced with this method I only obtain .inf as imitation loss, while if I save batches as output of a trining and I use them later for a supervised train loss is meaningful. I suppose there is a problem of format.

is there an automatic way to save some episodes given a custom policy?