Omar Nassar

Cybersecurity, A.I, and ML Enthusiast

View on GitHub

Machine Learning Projects

On this page, I will be uploading some machine learning projects that go through the process of loading data into a dataframe, inspecting the dataframe objects and attributes. Simply put, it will be summarizing my workflow methods of using different libraries such as sklearn, numpy, pyspark, and matplotlib for demonstrating my own point of view on how you can analyze, build, and eventually train a model. Confusion matrices will be used for means of measure, accordingly other measures of performances will be employed regarding the project’s nature.

KNN MultiClass Classification - Supervised ML

This is a fairly simple project, utilizing Sci-kit-learn's KNN classifier in a nested for loop to iterate through different N-Neighbors and different train/test subsetting. Label columns was Diabetes and had values of either 0, 1, or 2 thus indicating whether a patient was No-Diabetes, Type 1 Diabetic, or Type 2 Diabetic. The dataset was clean, and had no null-values so no pre-processing was done in this one.

Rendered Code Notebook | Dataset CSV File | Code Notebook