# Fine tuning segformer model

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Ffine-tuning-segformer-model%2F22715%3E%5C
**Category:** Beginners\
**Created:** [September 7, 2022, 10:52pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Ffine-tuning-segformer-model%2F22715 "2022-09-07T22:52:17Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![FahriBilici](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/fahribilici/32/5950_2.png) [@FahriBilici](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FFahriBilici)%5C
**Post date:** [September 7, 2022, 10:52pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Ffine-tuning-segformer-model%2F22715%2F1 "2022-09-07T22:52:17Z")

</div>

Hello, I am trying to train segformer model with my custom dataset. I have 46 different classes. My problem is my dataset labels are encoded pixel and there is multiple classes in one image. How I will generate the mask images for segformer like the example they share?