Keras or similar: Find correlation

I have a dataset consisting of close to 15,000 entries, each entry consisting of -1150 potential input variables (most of which are generally empty) and 10 target variables. The input for each entry is an ordered subset of said 1150 potential input variables, so any value given denotes the position within this subset. (Example: Lets say A, B, C and D are input variables. If we have CAD and no B, C would say 1, A 2, D 3, and B would be empty.) My hope is that a good correlation between input and target can be found.

Id like you to work with Python and Keras. Alternatively, please let me know which technology youd like to employ.

The truth is that I dont actually *know* whether or not a decent correlation can be established, but Id like you to try and see how good itll get. I tried myself, but my knowledge in this field is rather limited at best, and I wasnt successful. This also means that there are no clear criteria for accepance - Ill accept any code which looks reasonable. The flipside is that I see no reason for this to be a lengthy project at all if you know what youre doing, so be mindful of this when bidding.

Thank you!
About the recuiter
Member since Mar 14, 2020
Rehan Khanna
from Thuringia, Germany

---

---

Looking for help? Checkout our video tutorial
How to search and apply for jobs

How to apply? Do you have more questions about the Job?
See frequently asked questions