# Question about evaluation

**URL:** <https://discuss.ray.io/t/question-about-evaluation/12668>\
**Category:** RLlib\
**Created:** [October 31, 2023, 6:38pm UTC](https://discuss.ray.io/t/question-about-evaluation/12668 "2023-10-31T18:38:50Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![Destin](https://avatars.discourse-cdn.com/v4/letter/d/9e8a1a/32.png) [@Destin](https://discuss.ray.io/u/Destin)\
**Post date:** [October 31, 2023, 6:38pm UTC](https://discuss.ray.io/t/question-about-evaluation/12668/1 "2023-10-31T18:38:50Z")

</div>

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

- Low: It annoys or frustrates me for a moment.

I noticed that the Algorithm class provides an `[evaluate()](https://docs.ray.io/en/latest/rllib/package_ref/doc/ray.rllib.algorithms.algorithm_config.AlgorithmConfig.evaluation.html#ray.rllib.algorithms.algorithm_config.AlgorithmConfig.evaluation)` call function. My work flow is to perform training and then evaluation separately.

But the `evaluate()` function only works if I set the Algorithm’s evaluation config to have evaluation\_interval=1 meaning that it runs automatically after each call to train(). I current have evaluation\_parallel\_to\_training=False to prevent it from running concurrently with the train() command. But there does not seem to be any way to tell the algorithm to not perform evaluation until I call evaluate().

Is there a way to completely separate out the training from the evaluation processing?