# Any example of external errors for SameDiff?

**URL:** https://community.konduit.ai/t/any-example-of-external-errors-for-samediff/1896
**Category:** SameDiff
**Created:** [June 6, 2022, 5:01am UTC](https://community.konduit.ai/t/any-example-of-external-errors-for-samediff/1896 "2022-06-06T05:01:24Z")
**Posts on this page:** 1
**Page:** 1

<div class="post-metadata">

### Author: ![valb3r](https://yyz1.discourse-cdn.com/flex035/user_avatar/community.konduit.ai/valb3r/32/754_2.png) [@valb3r](https://community.konduit.ai/u/valb3r)
#### Post date: [June 6, 2022, 5:01am UTC](https://community.konduit.ai/t/any-example-of-external-errors-for-samediff/1896/1 "2022-06-06T05:01:24Z")

</div>

For DL4J there is a clear example of external error implementation [deeplearning4j-examples/MultiLayerNetworkExternalErrors.java at master · eclipse/deeplearning4j-examples · GitHub](https://github.com/eclipse/deeplearning4j-examples/blob/master/dl4j-examples/src/main/java/org/deeplearning4j/examples/advanced/features/externalerrors/MultiLayerNetworkExternalErrors.java)

It has nice property of having separate feed-forward and back-propagation steps that are public APIs. Is there any similar example for SameDiff, where feed-forward, back-propagation and model update steps are separate and are using public stable APIs?

My problem is that I need to perform network evaluation during training (as a step of network output evaluation) and for DL4J it is absolutely possible to do that using public APIs in a clear and concise way, but for SameDiff I’m lost as there is no explicit feed-forward, backdrop steps that do not modify internal states