# Dead link in parameter autotuning blog post

**URL:** <https://discourse.scverse.org/t/dead-link-in-parameter-autotuning-blog-post/88>\
**Category:** Site Feedback\
**Created:** [May 10, 2021, 2:39pm UTC](https://discourse.scverse.org/t/dead-link-in-parameter-autotuning-blog-post/88 "2021-05-10T14:39:13Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![Anton\_Bjork](https://yyz1.discourse-cdn.com/flex035/user_avatar/discourse.scverse.org/anton_bjork/32/45_2.png) [@Anton\_Bjork](https://discourse.scverse.org/u/Anton_Bjork)\
**Post date:** [May 10, 2021, 2:39pm UTC](https://discourse.scverse.org/t/dead-link-in-parameter-autotuning-blog-post/88/1 "2021-05-10T14:39:13Z")

</div>

I hope this is the right place.

This  
[https://yoseflab.github.io/2019/06/25/ZeroInflation/](https://yoseflab.github.io/2019/06/25/ZeroInflation/)  
link in the blog

> **[Hyperparameter search for scVI | scvi-tools](https://scvi-tools.org/blog/autotune/)**
>
> A study of the effects of hyperparameter optimization on scVI models.

about autotuning parameters is dead.  
Its linked from  
"  
See our group’s [blog post](https://yoseflab.github.io/2019/06/25/ZeroInflation/) by O. Clivio and P. Boyeau for a complete analysis of the conditional distribution choice  
"

I am guessing the link is supposed to be this one

> **[Should we zero-inflate scVI? | scvi-tools](https://scvi-tools.org/blog/zero-inflation/)**
>
> Droplet- based single-cell RNA sequencing (scRNA-seq) datasets typically contain at least 90% zero entries. How can we best model these zeros? Recent work focused on modeling zeros with a mixture of count distributions. The first component is meant...

---

<div class="post-metadata">

**Author:** ![adamgayoso](https://yyz1.discourse-cdn.com/flex035/user_avatar/discourse.scverse.org/adamgayoso/32/100_2.png) [@adamgayoso](https://discourse.scverse.org/u/adamgayoso)\
**Post date:** [May 10, 2021, 6:30pm UTC](https://discourse.scverse.org/t/dead-link-in-parameter-autotuning-blog-post/88/2 "2021-05-10T18:30:12Z")

</div>

Thanks! We will fix this soon.