We are currently living through a machine learning revolution and its applications are rapidly spreading to every aspect of our lives. Powering these applications are sophisticated statistical models. The excellent Python machine learning library Scikit-Learn provides access to these powerful models so that we can quickly begin applying them to our data. In this two-day machine learning training course with Python you will learn the skills to apply machine learning models to your datasets.
Key Learning Areas
After this Dcourse completion, you will have the skills to apply machine learning models to your datasets. You will become intimately familiar with how the Scikit-Learn API works and how to make the most of it.
You will learn:
- Preparation of data for machine learning
- The Scikit-Learn API
- Supervised vs Unsupervised Learning
- Model training, tuning, validating, and predicting
- Visualization of validation metrics
- Building machine learning pipelines
This a very hands-on course and students should expect to spend at least half of class time coding. All material will be contained in Jupyter Notebooks.
Data Preparation for Machine Learning
- The data science lifecycle
- Exploratory data analysis
- Idiosyncrasies of Scikit-Learn
- Estimators vs. Functions
- Handling missing data
- Non-numeric data
Practical Machine Learning
- Building a dummy model
- Training and predictions
- Model validation
This course is for those who are interested in practical application of machine learning models using the Python machine learning library Scikit-Learn. The entire focus of the course will be on programmatically training and validating models on real-life datasets.
This course assumes you know the basic fundamentals of Python.