This repo contains all the files needed to deliver the Data Science On-Ramp course. It was originally delivered as a 3-hour session to the Northern Virginia Professional Association of SQL Server (PASS), but the content can easily fill a full day or more depending on how much time is dedicated to hands-on lab work. To accompany the lab files this repo contains three (3) PowerPoint presentations:
- The first presentation covers the fundamental statistical concepts, techniques and measures involved in performing statistical data analysis. These include descriptive statistics, probability theory, hypothesis testing and inferential statistics. It goes on to detail the various tests for handling both parametric and non-parametric distributions, as well as for handling categorical data.
- The second presentation covers machine learning. It describes how the fundamental statistical methods are executed for the purpose of rendering predictions based on data samples. In doing so it addresses both supervised and unsupervised learning methods including classification, regression, clustering, and dimensionality reduction using Principle Component Analysis (PCA). The presentation closes with a discussion of Deep Learning with a focus on using Convolutional Neural Networks (CNN) to train image analysis solutions.
- The third presentation covers SQL Server 2017 Machine Learning Services which provides a means for operationalizing machine learning models and efficiently conducting machine learning development at scale, securely and in a highly performing environment.
The lab files for this course correspond to the PowerPoint presentations, and are implemented using the Jupyter Notebooks IDE using the Python programming language including the use of numerous popular data science libraries; e.g., Numpy, Pandas, Matplotlib, and Scikit-Learn.