Data Science with Python

Data Science with Python

Description

This course is an introduction to Machine Learning and Optimization with Python. The course starts defining mathematical modelling and how it can be used in real business cases. Examples of supervised and unsupervised learning are provided with a presentation of the main methodologies to train performant models. Then the course completes the picture with a deep dive into time series analysis and optimization problems. The course is designed to be interactive and practical. The student will learn by doing, by solving real business problems.

Topics include

1. The present and future of Data Science
2. Technologies and integration with Python
3. Concepts of AI, machine learning, and statistical learning and the role of Python
4. Supervised learning: Linear models, Bayesian classifiers, Random forests, Xgboost
5. Unsupervised learning: K-means, hierarchical clustering, dimensionality reduction algorithms
6. Introduction to Scikit-learn
7. Model evaluation and improvement: cross-validation, k-folds, grid search, key metrics
8. Chaining algorithms and creating pipelines
9. Introduction to the Statsmodels library
10. Integration of Python with enterprise frameworks and tools
11. Time series analysis
12. Time series modeling: ARMA, ARIMA, SARIMA
13. Time series forecasting
14. Time series classification
15. Time series clustering
16. Fourier Transform and signal analysis
17. Illustrative examples of other approaches to using time series data
18. Optimization algorithms and libraries
19. Local Vs Global optimization
20. Linear programming
21. Quadratic programming
22. Nonlinear optimization

What you will be able to do

1. Understand the main concepts of Data Science
2. Use the main libraries for Data Science in Python
3. Understand the main algorithms for supervised and unsupervised learning
4. Evaluate and improve models
5. Create pipelines for chaining algorithms
6. Use the Statsmodels library
7. Integrate Python with enterprise frameworks and tools
8. Analyze time series data
9. Model time series data
10. Solve optimization problems with or without constraints with the most efficient methods

Duration

3 days

Prerequisites

Ability to use Python for data analysis. A mathematical background is recommended.

Audience

This is an intermediate course for professionals and researchers who need to solve business problems with data.