# External data | Pre-train models
**URL:** https://community.drivendata.org/t/external-data-pre-train-models/7693
**Category:** Where's Whale-do?
**Created:** [May 5, 2022, 6:02pm UTC](https://community.drivendata.org/t/external-data-pre-train-models/7693 "2022-05-05T18:02:36Z")
**Posts on this page:** 10
**Page:** 1
### Author:  [@radmirkaz](https://community.drivendata.org/u/radmirkaz)
#### Post date: [May 5, 2022, 6:02pm UTC](https://community.drivendata.org/t/external-data-pre-train-models/7693/1 "2022-05-05T18:02:36Z")
It’s a bit strange that we should share external data and pre-train models. But do you guys have some resources?🙂
---
### Author:  [@Infin](https://community.drivendata.org/u/Infin)
#### Post date: [May 16, 2022, 5:10pm UTC](https://community.drivendata.org/t/external-data-pre-train-models/7693/2 "2022-05-16T17:10:46Z")
## i am interested too
---
### Author:  [@\_NQ](https://community.drivendata.org/u/_NQ)
#### Post date: [June 25, 2022, 5:02am UTC](https://community.drivendata.org/t/external-data-pre-train-models/7693/3 "2022-06-25T05:02:12Z")
Disclosing [Multimedia Commons Dataset](http://multimedia-commons.s3-website-us-west-2.amazonaws.com/?prefix=)
Licenses available in: prefix=tools/etc/yfcc100m\_dataset.sql (CC0 and CC-BY images only)
> **[1503.01817.pdf](https://arxiv.org/pdf/1503.01817.pdf)**
>
> 4.17 MB
These are publicly acessible with the “UNSIGNED” token, see e.g. [python - Can I use boto3 anonymously? - Stack Overflow](https://stackoverflow.com/questions/34865927/can-i-use-boto3-anonymously)
* * *
Also:
> **[GitHub - cvdfoundation/google-landmark: Dataset with 5 million images...](https://github.com/cvdfoundation/google-landmark)**
>
> Dataset with 5 million images depicting human-made and natural landmarks spanning 200 thousand classes. - GitHub - cvdfoundation/google-landmark: Dataset with 5 million images depicting human-made ...
[https://storage.googleapis.com/openimages/web/index.html](https://storage.googleapis.com/openimages/web/index.html)
[https://www.robots.ox.ac.uk/~vgg/data/pass/](https://www.robots.ox.ac.uk/~vgg/data/pass/)
---
### Author:  [@sheep](https://community.drivendata.org/u/sheep)
#### Post date: [June 29, 2022, 7:22am UTC](https://community.drivendata.org/t/external-data-pre-train-models/7693/4 "2022-06-29T07:22:12Z")
tf\_efficientnet\_bN\_ns, swin and convnext imagenet pretrained from timm package
---
### Author:  [@Ammarali32](https://community.drivendata.org/u/Ammarali32)
#### Post date: [June 30, 2022, 8:06am UTC](https://community.drivendata.org/t/external-data-pre-train-models/7693/5 "2022-06-30T08:06:56Z")
From my side:
No external data were used.
tf\_efficientnet\_b5\_ns (noisy-student), efficientnetv2\_rw\_m pretrained from timm package
---
### Author:  [@adityakumarsinha](https://community.drivendata.org/u/adityakumarsinha)
#### Post date: [June 30, 2022, 2:31pm UTC](https://community.drivendata.org/t/external-data-pre-train-models/7693/6 "2022-06-30T14:31:23Z")
timm pretrained models: tf\_efficientnet\_b\*_ns, convnext __, eca\_nfnet\__
no external data
---
### Author:  [@karelds](https://community.drivendata.org/u/karelds)
#### Post date: [June 30, 2022, 7:07pm UTC](https://community.drivendata.org/t/external-data-pre-train-models/7693/7 "2022-06-30T19:07:10Z")
Already declared but just in case: Pre-trained _EfficientNet_ models from _timm_
---
### Author:  [@flamethrower](https://community.drivendata.org/u/flamethrower)
#### Post date: [June 30, 2022, 7:40pm UTC](https://community.drivendata.org/t/external-data-pre-train-models/7693/8 "2022-06-30T19:40:28Z")
Pre-trained imagenet- TF\_EfficientNet, ConvNext, ECA\_NFNET, RegNetZ models all from timm.
---
### Author:  [@\_NQ](https://community.drivendata.org/u/_NQ)
#### Post date: [June 30, 2022, 10:43pm UTC](https://community.drivendata.org/t/external-data-pre-train-models/7693/9 "2022-06-30T22:43:23Z")
Assume this isn’t required, but efficientnetv2, seresnet152d, ecaresnet269, and more (timm==0.5.4)
---
### Author:  [@qwerty64](https://community.drivendata.org/u/qwerty64)
#### Post date: [July 11, 2022, 4:51pm UTC](https://community.drivendata.org/t/external-data-pre-train-models/7693/10 "2022-07-11T16:51:29Z")
timm pretrained models: tf\_efficientnet\_b\*ns
no external data