Skip to content
Switch branches/tags

Latest commit


Git stats


Failed to load latest commit information.
Latest commit message
Commit time


Unicorn analytics retrieves analytics data from the YouTube Analytics API and Google Analytics API. The unicorn in the story is the unicorn hat HD. Page hits and video views are visualized as a snake crawling over the unicorn display. One snake equals one page load or view.

If you don't have that many views you can set a scaling factor - for example we can retrieve the statistics over the last week; and display them over 5 minutes - then refresh the statistics and play them again for 5 minutes.

sample image from internet

GIF of me running the unicorn-analytics on a Raspberry PI 3, replaying page loads and youtube views for the last week over 5 minutes. I used the configuration that is checked in here: unicorn-analytics/config

Related project: pi-zero-ethermeter


You need at least one raspberry pi with a unicorn HAT. The raspberry pi needs python 3.5+ to support await/async.

Modules that needs to be installed,

pip3 install oauth2client aiohttp
pip3 install unicornhathd numpy
  • aiohttp: to asynchronously update the statistics without disturbing the animations.
  • oauth2client: used to retrieve oauth2 tokens from gooooogle.
  • httplib2: to update the access token when it expires its a bit messy so we didn't implement this with aiohttp.

Now is a good time to configure your google APIs, see the section on Configuration.

To start it all up run,

git clone
cd unicorn-analytics
python -m

Make sure you have configured everything first 😏

The first time it is launched it will ask you to open your browser and allow third party access. If you don't have a browser available on your PI, please run as follows:

python -m --noauth_local_webserver

And open your browser on another machine.

Supported API's

  • Google Analytics
  • YouTube Analytics


Perform the following and update config/analytics.json and config/youtube.json

  • create a new Google Analytics project, grab the view ID.
  • make sure there is a YouTube channel registered on your Google account.

Create a new API project on the google developers console

Create an OAUTH2 client secret by following these instructions;

Download it as json and place under api_keys/analytics_secret.json and api_keys/youtube_secret.json.

Additional notes

For the youtube analytics API we may only authenticate using a OAUTH2 token. Most of the examples in the docs assume that we are a web application. Some hours later I actually found out that there IS a flow that can be used from the commandline! Scroll to the bottom of the following page:

With this we can actually generate oauth2 tokens without requiring a browser to be opened on the PI. If you don't have a browser available in your environment please pass the commandline parameter



All contributions welcome. 🐟 💦


retrieve analytics from google APIs and visualize on a unicorn hat - asyncio/aiohttp








No releases published


No packages published