TAMU Datathon - Learner Track

Lecture 1: Data Basics

• The basic vocabulary of data:
• Tabled Data vs Unstructured Data
• Features and Target Columns
• Samples (data rows)
• The basics of reading/loading data using Pandas
• Deal with missing values
• Plots: probability distribution and correlation (heatmaps) plots
• How to visualize data and understand it better:
• Exploratory analysis
• Communicate effectively
• How to make recommendations to stakeholders

Lecture 2: Linear Regression

• What is a model? What makes a model linear?
• Fit your own model by hand!
• What does it mean to fit a model? (What is machine learning?)
• A visual introduction to optimization - "What is a Loss Function?"
• How to train a linear model using the Scikit Learn library
• How to interpret a linear regression model
• Understand the interpretability <> accuracy tradeoff

Lecture 3: Logistic Regression

• What is classification?
• What is logistic regression?
• A visual introduction to non-convex optimization
• Fit a logistic regression model yourself!
• Fit a logistic regression model using the Scikit Learn library
• Fake News!! What is accuracy, true/false positives/negatives, precision, recall?
• Generalize logistic to multiple classes
• Visualize 2D decision boundaries

15 hours of hacking time to work on a data science project to solve a real-life challenge.