search

jeffheaton / t81_558_deep_learning

T81-558: Keras - Applications of Deep Neural Networks @Washington University in St. Louis
https://sites.wustl.edu/jeffheaton/t81-558/
Other
5.72k stars 3.03k forks source link

Z-score formula is wrong in t81_558_class02_python_ml.ipynb #21

Closed alpmarangoz closed 5 years ago

alpmarangoz commented 5 years ago

Hello Jeff, thank you very much for the lecture. There is a small typo in Z-score formula in t81_558_class02_python_ml.ipynb / Section Field Transformation & Preprocessing. It should be equation instead of equation

jeffheaton commented 5 years ago

Thanks, yes that makes sense, I will update.

jeffheaton commented 5 years ago

I can't find the formula that you are referring to. If it is a single-line equation, like you have there, I agree that ()'s are needed. But the formula that I have in the course uses a ratio, so it is x - mu OVER sigma. Therefore the grouping between x and rho is implied. Wikipedia has the formula in the same form.

alpmarangoz commented 5 years ago

It is a typo in the text part of the Jupyter notebook. Just before “In (9):”, under the heading “Field Transformation & Preprocessing”, you explain the Z-Score formulation. The formula you give should be corrected. It is a minor thing but I thought your students might get confused. Best regards.

jeffheaton commented 5 years ago

Okay fixed it. That was strange, it displayed correctly in Jupyter Notebook, but not GitHub. It seems the LaTex \over is handled differently by the GitHub viewer. I switched to \frac and it seems to display correctly on both. Thanks for pointing that out!