# Disclosure of top solution

**URL:** <https://community.drivendata.org/t/disclosure-of-top-solution/10360>\
**Category:** Kelp Wanted: Segmenting Kelp Forests\
**Created:** [February 22, 2024, 3:25pm UTC](https://community.drivendata.org/t/disclosure-of-top-solution/10360 "2024-02-22T15:25:38Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![hengcherkeng](https://avatars.discourse-cdn.com/v4/letter/h/85e7bf/32.png) [@hengcherkeng](https://community.drivendata.org/u/hengcherkeng)\
**Post date:** [February 22, 2024, 3:25pm UTC](https://community.drivendata.org/t/disclosure-of-top-solution/10360/1 "2024-02-22T15:25:38Z")

</div>

congratulations to all winners and top scorers!

I am very interested to know how to get the private dice score in the 0.73+ and 0.72 range.  
is there any tricks in the segmentation models?

i only manged to get 0.72+ in my local cv and 0.71+ for private scores.

---

<div class="post-metadata">

**Author:** ![hjdeheer](https://yyz2.discourse-cdn.com/flex028/user_avatar/community.drivendata.org/hjdeheer/32/1656_2.png) [@hjdeheer](https://community.drivendata.org/u/hjdeheer)\
**Post date:** [February 22, 2024, 3:39pm UTC](https://community.drivendata.org/t/disclosure-of-top-solution/10360/2 "2024-02-22T15:39:56Z")

</div>

Hi,

We will publish our code and a full report of our solution later today! We are also looking forward to others’ people solutions 🙂

Kind regards,

Team Epoch IV