Course Overview

Python in Data Science is huge right now. This course covers the basics to get started looking with classification and regression in Machine Learning. We will use the popular Jupyter and scikit-learn tools to create models, train them, evaluate them, and finally interpret the results. We will use standard numeric evaluators as well as visualization libraries to help us. There are ample labs for you to try out the tools on your own and gain mastery with them.

Topics covered include using creating baseline models, model families, evaluation, hyperparameter optimization, and interpreting models, and more. Taking this course will teach you how to leverage the scikit-learn library and various supporting libraries like Lime and Yellowbrick.

* This course can run two to four days (extra days for NLP/text analysis and unsupervised learning with PCA/Clustering).

Key Learning Areas

  • Classification overview
  • Baseline Model
  • ROC
  • Model Interpretation
  • Hyperparameter Optimization
  • Evaluating data quantity
  • Other evaluation metrics (Accuracy, recall, precision, F1)
  • Precision Recall Curves
  • Class balance
  • *Text classification
  • Boosting technique
  • ML Algorithm Families
  • Regression overview
  • Regression evaluation
  • * Unsupervised Learning (PCA, clustering)

Course Outline

This course will cover the skills to be comfortable with classification and regression using scikit-learn and Python. With hands-on labs, you will be able to quickly evaluate your mastery of the topics. We don’t just show how to use the library, but how to improve the model, evaluate the results, and interpret the findings. The topics are presented with real world data. Finally, there are many labs to make sure that you get the chance to try them out.

Python is wonderful basis to leverage for Machine Learning. We will explore the scikit-learn library so you can create effective predictive models. You will evaluate the models using numerical values, as well as visualizations with the Yellowbrick library. If you intend to get into Machine Learning, this is an excellent course, as it focuses on the popular libraries in Python.

Who Benefits

This course is for experienced Python developers or analysts who have some experience with Python. Typical attendees are looking to leverage Python to create predictive models.

If you want to leverage Python tor Machine Learning, this course will get you started. If you are on a team using Python and pandas, this will be a great opportunity to get your team on the same page and speaking the same language.

Prerequisites

This course assumes the students have solid working knowledge of the Python language. They should also have some experience with the pandas library.