Bug: Normalizer Z-Trans needs to be reviewed

[detlefarend]

1 Describe the bug

• Class bf.math.normalizers.NormalizerZTrans
  • Method update_parameters() is not prepared for just remove a single data entity. If just p_data_del is filled an error is raised.
  • Method update_parameters(): equations are hard to recap and maybe wrong

2 To Reproduce Steps to reproduce the behavior:

1. Go to '...'
2. Click on '....'
3. Scroll down to '....'
4. See error

3 Expected behavior A clear and concise description of what you expected to happen.

4 Screenshots If applicable, add screenshots to help explain your problem.

5 Additional Informations 5.1 Operating System

• [ ] Linux
• [ ] Windows
• [ ] Mac OS

5.2 Branch bf/oa/streams/refact

6 References http://datagenetics.com/blog/november22017/index.html

[laxmikantbaheti]

Hi @detlefarend I am working on this. I will give you update

[laxmikantbaheti]

@detlefarend , just a short update, the version of z transformation from 30.04.2024 works perfect for both new and deleted data. I need to verify changes after that. I am on it...

[laxmikantbaheti]

Hi @detlefarend , I have fixed a small bug in the update parameter method. I like how you seperated the method for both cases, new and old data, it will automatically make sure all the possible cases. There was just a small bug in the equation. Now it's fixed, by seeing the howto_bf_math_011. Just needed some fix in the means and std equations.

To understand the logic of updating parameters on p_data_del: The calculation of STD for a new value was already correct. i.e. we could calulate std_new based on old mean and std. To update the std on deleted data, we can think of it also as _migrating back from the std of dataset with added data to the one without the new data (the new data is basically datadel in this case). And then just solve the equation.

Let me know whenever you are able to take a look.