100 Days of ML Code – Day 1

So, I finally started 100 days of ml code! This is a github repo I stumbled upon some time some where and I have forked it in order to beeing able to document my journey. The original repo is here.
After going through the notes for Day 1 I realized that the seven years, the guide is old, migt become a problem. For example the python libraries used for the lessons (Pandas, scikit-learn and NumPy) have evolved. And usage of the used classes and functions is no longer valid. However feeding my code and error messages into Google Gemini, I did not have to be a connoisseur of Pandas, NumPy and scikit-learn. Gemini explained to me, how to update the out-dated code (Updated code for Day 1). After having overcome this hurdle, my takeaway from Day 1 is like this:

• Loading csv data with pandas read_csv
• Using scikit-learn SimpleImputer class to make sure missing data is represented correctly
• Transform categorical data with scikit-learn LabelEncoder
• Feature Scaling with scikit-learn StandardScaler, making sure high and low magnitudes are equally weighted

In a nutshell: Read a csv file plus some machine learning „bells and whistles“ 🙂