Lesson 3 Rossman applied to horse racing data

Hi all,

One of my pet projects has been the collection and analyzing of horses racing data.

I have used the Rossman Lesson 3 Structured Data as a foundation.

I have shared the data on Kaggle and my notebooks on Github.

Haven’t been able to crack the code but has been really useful in really understanding how structured data for neural nets works.

Hope someone can get some value from it.




Hi @lukebyrne let me know if you’re still around and interested in discussing this further

Hi Gumby, I am still around, I have progressed this project to this stage, https://oddsrepublic.com

Hit me up to discuss.

Thanks for replying.
I’m on your new project now - will monitor throughout today and see how it goes. I’m sure you’re across the duplicate horses for each race I saw during my cursory view. Have you been logging your results? How’s it going?

I’ve been struggling to setup a Machine Learning horse racing prediction project for a number of years. My view is you need more than just basic form parameters (i.e. barriers, 1st up, 2nd up etc) to produce something meaningful (i.e. time, tempo, benchmark ratings).

I have a sample raw data file that I’ve put together, but I don’t understanding from a labelling perspective what I need to do to get it working. Was wondering if you would be interested in taking a look.

Hi Gumby, yeh there are dupes on Morphetville.

Ping me over your data file and I can take a look.