Numerai dashboarding and data access, part 1 (January 30, 2020)


I use text-based dashboards reporting stocks/crypto buy/sell opportunities and Numerai forecasting performance. Why text-based? It's fast to generate and works great with Notepad++'s refresh function. Having dashboards with the exact information I want beats having to log into various websites and do a bunch of clicking. Also, I don't need pretty for personal use.

Update. 20200816. The version here no longer works well due to some backend changes from Numerai. You can find an working version of this here.

You can find my Numerai dashboard code here (link) as a Colab notebook. You can run the first code block to get a feel for how things work. You can review the code and notes in the second code block to get it setup in Windows. I use Python on my computer, and I task schedule a batch file to run daily to update the data. You will need to pip install the needed APIs for this to work outside of Colab. My setup consists of a "Daily Report.bat" that the task scheduler calls to run a py file via Winpython. The py file generates a txt file that Notepad++ refreshes. Here's what the dashboard looks like:

Picture showing the output generated by the Colab/Python dashboard code.

           

The code uses the numeraiAPI build by uuazed to replicate some of the data you would find on the Numerai Tournament status page. While you might prefer to just go to the website, the notebook shows how you can access the data in case you change your mind down the road. Due to the Numerai site being attacked recently the security has upped and login has become a hassle. 

The Numerai website shows the data in specific formats and orders that you might not like. You might also want to get access to the historical data that's not shown on the website, I will have a Part 2 later showing how you can access and plot those data. 

Part 2 verifying Numerai submission
Part 3 dashboarding the daily performance data