# How to add custom layers for a pytorch network?

**URL:** <https://forums.developer.nvidia.com/t/how-to-add-custom-layers-for-a-pytorch-network/81802>\
**Category:** TensorRT\
**Created:** [September 17, 2019, 3:01am UTC](https://forums.developer.nvidia.com/t/how-to-add-custom-layers-for-a-pytorch-network/81802 "2019-09-17T03:01:45Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![J-Penny](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/j-penny/32/11637_2.png) [@J-Penny](https://forums.developer.nvidia.com/u/J-Penny)\
**Post date:** [September 17, 2019, 3:01am UTC](https://forums.developer.nvidia.com/t/how-to-add-custom-layers-for-a-pytorch-network/81802/1 "2019-09-17T03:01:45Z")

</div>

Platform：Jetson Xavier  
TensorRT version: 5.0.6.3  
CUDA version: 10.0

Hello，  
I want to use use TensorRT with PyTorch, but some layers such as upsamples that are not supported.Can I use custom layers to implement them? If so ,what should I do to implement them?In addition, whether the tensorrt version I am currently using can support this operation?

Looking forward to your reply, thanks .