# Working with models and data

**URL:** <https://community.drivendata.org/t/working-with-models-and-data/4509>\
**Category:** Hateful Memes\
**Created:** [May 24, 2020, 7:54pm UTC](https://community.drivendata.org/t/working-with-models-and-data/4509 "2020-05-24T19:54:36Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![dn298](https://avatars.discourse-cdn.com/v4/letter/d/2bfe46/32.png) [@dn298](https://community.drivendata.org/u/dn298)\
**Post date:** [May 24, 2020, 7:54pm UTC](https://community.drivendata.org/t/working-with-models-and-data/4509/1 "2020-05-24T19:54:37Z")

</div>

Hey there!

My team and I tried to get some benchmark scores using LoRRA, Pythia, and other pretrained models, but could get them up and running. Below is a picture of the error.

![Screen Shot 2020-05-24 at 3.53.10 PM](https://canada1.discourse-cdn.com/flex028/uploads/drivendata1/original/1X/d17c984c6a54ca507ff15e82fd21fdc53732f059.png)

Should we be downloading more files and data than what is given in the challenge? I downloaded the data from the challenge site, and then followed these install instructions after forking the repo:

[https://mmf.readthedocs.io/en/latest/notes/installation.html](https://mmf.readthedocs.io/en/latest/notes/installation.html)

Thanks!

---

<div class="post-metadata">

**Author:** ![amanpreet](https://avatars.discourse-cdn.com/v4/letter/a/49beb7/32.png) [@amanpreet](https://community.drivendata.org/u/amanpreet)\
**Post date:** [May 25, 2020, 7:32am UTC](https://community.drivendata.org/t/working-with-models-and-data/4509/2 "2020-05-25T07:32:31Z")

</div>

Hi,

MMF creator here. Please open up MMF related issues on [MMF](https://github.com/facebookresearch/mmf/issues/new/choose) directly.

Thanks.