Feature: try out different models and training strategies using notebooks and document them

[KarinePistili]

We would like to test out different AI models and methods for the eye tracking process. We ask you to save your test and findings on notebooks inside the repository. The idea is to make simple proof of concepts so we can comparte the different methods.

[gitone912]

can i do this one , i am good at modelling

[KarinePistili]

Yes you can. I'll assign it you @gitone912 . If you have any questions feel free to send them here ;)

[gitone912]

sure thanks a lot @KarinePistili

[gitone912]

hey @KarinePistili i tried different models on the data we generate through calibration process , the accuracy and error rates are all depends on how perfect the user calibrate so their data dispersion becomes less.
the question arises here in my mind is what exact goal are we trying to accomplish using different models because more or less accuracy and errors are dependent on how user interact in test so even if we change various model , superiority of data will always be high
please clear my doubt
thankyou