# Building a COI database from BOLD references

**URL:** https://forum.qiime2.org/t/building-a-coi-database-from-bold-references/16129
**Category:** Tutorials
**Created:** [August 14, 2020, 10:43pm UTC](https://forum.qiime2.org/t/building-a-coi-database-from-bold-references/16129 "2020-08-14T22:43:29Z")
**Posts on this page:** 1
**Showing post:** 2

<div class="post-metadata">

### Author: ![Margreet](https://forum.qiime2.org/letter_avatar_proxy/v4/letter/m/13edae/32.png) [@Margreet](https://forum.qiime2.org/u/Margreet)
#### Post date: [August 27, 2020, 8:16am UTC](https://forum.qiime2.org/t/building-a-coi-database-from-bold-references/16129/2 "2020-08-27T08:16:24Z")

</div>

This is amazing! Thank you so much for all the effort, explanations and sharing your ideas with us. I am really excited.

At this moment, I am trying to produce a classifier for the bold\_derep1 dataset. I am trying this with the help of the qiime rescript evaluate-fit-classifier as indicated at the end of the tutorial. However, I am struggling with the memory errors I am getting over and over again:

"Unable to allocate 11.8 GiB for an array with shape (8192, 192871) and data type float64".

I am using VirtualBox, which is already has a Base memory of 53091 MB.. Does anyone know how I can improve this process so the classifier can be created?

Thanks in advance!

Margreet

---

_[View the full topic](https://forum.qiime2.org/t/building-a-coi-database-from-bold-references/16129)._