Docs: Incorrect FP/FN definitions in the Confusion Matrix explanation for Logistic Regression

[czn1572179629]

Hi team,

First off, thank you for the excellent documentation on this project. It's been very helpful.

I was studying the Logistic Regression section, specifically the part explaining the Confusion Matrix, and I believe there's an error in the definitions provided for False Positive (FP) and False Negative (FN). The current explanation seems to have them reversed.

1. Location of the error:
File: 2-Regression/4-Logistic/README.md

2. The current (incorrect) text:
If your model predicts a pumpkin as white and it belongs to category 'not-white' in reality we call it a false negative...
If your model predicts a pumpkin as not white and it belongs to category 'white' in reality we call it a false positive...

3. The core issue:
The text incorrectly defines a False Positive as a miss and a False Negative as a false alarm. The standard definitions are the other way around.

4. Suggested Correction:

For clarity, here are the standard definitions, assuming "white" is the positive class:

• False Positive (FP): The model predicts Positive ("white"), but the actual class is Negative ("not-white"). (A "false alarm")
• False Negative (FN): The model predicts Negative ("not-white"), but the actual class is Positive ("white"). (A "miss")

Based on this, the documentation could be updated to:

If your model predicts a pumpkin as white (Positive) and it belongs to the category 'not-white' (Negative) in reality, we call it a false positive.

If your model predicts a pumpkin as not white (Negative) and it belongs to the category 'white' (Positive) in reality, we call it a false negative.

I hope this feedback is helpful. I'd also be happy to submit a Pull Request to fix this if you'd like.

Thanks again!

[czn1572179629]

Duplicate of #825 . Closing.