Udacity Nanodegree: Data Science

For term 1, it containes 3 parts: Supervised, Deep, and Unsupervised Learning

The Exercises are included in each folder.

The Projects are also included in their own folders:

1. Supervised Learning: finding charity donors, extracted feature importance, using data collected from the 1994 U.S. Census (Naive Bayes, Random Forest, and SVM)

2. Deep Learning: Flower Species Image Classifier: Designed a flower classifier with PyTorch (ResNet, VGG, CUDA, Command Line App); Achieved 90% accuracy, performed sanity check with visualization (Matplotlib))

3. Unsupervised Learning: Customer Identification for Mail-Order Sales Company: Identified core customers by interpreting the differences between the clusters for the general population and that of customers (PCA, K-Means); Built pipeline includes data cleaning, feature engineering, modeling, clustering