# 'timesteps\_per\_iteration' parameter

**URL:** <https://discuss.ray.io/t/timesteps-per-iteration-parameter/2531>\
**Category:** RLlib\
**Created:** [June 15, 2021, 10:44pm UTC](https://discuss.ray.io/t/timesteps-per-iteration-parameter/2531 "2021-06-15T22:44:32Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![Fot](https://avatars.discourse-cdn.com/v4/letter/f/b2d939/32.png) [@Fot](https://discuss.ray.io/u/Fot)\
**Post date:** [June 15, 2021, 10:44pm UTC](https://discuss.ray.io/t/timesteps-per-iteration-parameter/2531/1 "2021-06-15T22:44:32Z")

</div>

Hi all! I am a bit confused about the role of the _timesteps\_per\_iteration_ parameter, and how is it related to the _train\_batch\_size_ parameter. Could someone please clarify their differences?

---

<div class="post-metadata">

**Author:** ![SebastianBo1995](https://avatars.discourse-cdn.com/v4/letter/s/e9bcb4/32.png) [@SebastianBo1995](https://discuss.ray.io/u/SebastianBo1995)\
**Post date:** [July 21, 2021, 7:26am UTC](https://discuss.ray.io/t/timesteps-per-iteration-parameter/2531/2 "2021-07-21T07:26:10Z")

</div>

Hi,

I also wanted top knownwhat it is exactly doing.  
Just a system Parameter:

> <https://github.com/ray-project/ray/issues/4506>
>
> Hi, 
> 
> I wanted to know more about what the \`timesteps\_per\_iteration\` parameter… means in context of DQN based agents? Quick glance at the code: https://github.com/ray-project/ray/blob/cff08e19ff1606ef6e718624703e8e0da19b223d/python/ray/rllib/agents/dqn/dqn.py#L257-L261 
> suggests that you optimize for \`timesteps\_per\_iteration\` after each env step? That doesn't quite seem right since the default value for \`timesteps\_per\_iteration\` is 1000 which seems high . 
> 
> Also, does \`timesteps\_per\_iteration\` have a different connotation in case of distributed agents like ApeX?