# Compute resources for the competition?

**URL:** <https://community.drivendata.org/t/compute-resources-for-the-competition/3779>\
**Category:** Mapping Disaster Risk from Aerial Imagery\
**Created:** [October 30, 2019, 8:41pm UTC](https://community.drivendata.org/t/compute-resources-for-the-competition/3779 "2019-10-30T20:41:37Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![krishnab75](https://avatars.discourse-cdn.com/v4/letter/k/9dc877/32.png) [@krishnab75](https://community.drivendata.org/u/krishnab75)\
**Post date:** [October 30, 2019, 8:41pm UTC](https://community.drivendata.org/t/compute-resources-for-the-competition/3779/1 "2019-10-30T20:41:37Z")

</div>

Say, I was just wondering if there are any compute resources available for this competition? I am talking about AWS, GCP, or other resources to train the models. Vision models take such a long time, that it would be nice to access some TPUs or V100s when models are ready to ramp up.

---

<div class="post-metadata">

**Author:** ![bhali12](https://avatars.discourse-cdn.com/v4/letter/b/9dc877/32.png) [@bhali12](https://community.drivendata.org/u/bhali12)\
**Post date:** [November 21, 2019, 3:56pm UTC](https://community.drivendata.org/t/compute-resources-for-the-competition/3779/2 "2019-11-21T15:56:01Z")

</div>

You can use google colab! They have free GPU fir this type of problems.

> **[Google Colab: Jupyter Lab on steroids (perfect for Deep Learning)](https://towardsdatascience.com/google-colab-jupyter-lab-on-steroids-perfect-for-deep-learning-cdddc174d77a)**
>
> Plus cute animated corgies and cats on your notebook

---

<div class="post-metadata">

**Author:** ![bkowshik](https://yyz2.discourse-cdn.com/flex028/user_avatar/community.drivendata.org/bkowshik/32/880_2.png) [@bkowshik](https://community.drivendata.org/u/bkowshik)\
**Post date:** [November 30, 2019, 6:59pm UTC](https://community.drivendata.org/t/compute-resources-for-the-competition/3779/3 "2019-11-30T18:59:12Z")

</div>

Just got up and running with Google Colab and wrote down the steps at the link below:

> [@Requesting GCP credits](http://community.drivendata.org/t/requesting-gcp-credits/3811/2):
>
> wave @sushant3095 I was able to get up and running for free on Google Colaboratory in about 30 mins with the following steps. Visit [https://colab.research.google.com/](https://colab.research.google.com/) and create a new Python 3 notebook In the Runtime menu, select Change runtime type and then select Hardware accelerator as GPU Download the entire dataset of ~30GB using wget wget https://s3.amazonaws.com/drivendata-public-assets/stac.tar You will find the dataset available at the path, /content/stac.tar in about 15 mins …