Training: Data science with Python

The basic training in Python with the focus on data science is the ideal introduction for analysts to the programming language. In addition to programming paradigms, object types and syntax structures and pandas, the central library for data management. Through the training the participants will be able to implement the data management as well as the first analyses and visualizations in Python.

Table of contents:

• First steps into Python
• Concept and philosophy of Python
• Data structures and their properties
• Importing data
• Data management with pandas
• Data analysis with Python
• Loops and control elements

—

Recommended course length: Two days

The aim of the course is to learn machine learning methods by applying them to practice-oriented exercise data sets. During the training, participants will learn central steps such as the preparatory data management, the training of algorithms as well as the forecasting, validation and implementation of them directly in Python. The special focus of the course lies on the Python library scikit-learn, which contains a variety of popular algorithms in the field of machine learning.

During the course, participants will create Python scripts which can be used as templates for their own machine learning applications.

Table of contents:

• Basics of machine learning / data mining
  • Overview of models and methods, of the problem of forecasting
  • Supervised vs. unsupervised learning
• Overfitting, underfitting and parameter tuning – modeling techniques
  • Basic problem, simple cross-validation, 3-fold technique, k-fold validation, cross-validation for time series
• Classification methods
  • Decision trees, random forest, gradient boosting machines, neural networks, support vector machines
  • Evaluation of classification procedures (ROC curves, cut-off value, precision, sensitivity, specificity)
• Regression problems
  • Linear regression, regression trees, neural networks, support vector machines, random forest
  • Evaluation of regression problems
• Cluster analysis
  • k-nearest neighbors, k-means, agglomerative cluster analysis, visualizations

—

Recommended course length: Two days

With Python, patterns and structures in data with time dimensions can be recognized and thus reliable forecasts can successfully derive from them. By using practice-oriented data sets, you will learn these analysis steps in our course on time series analysis with Python.

We will introduce you to the background of time series analysis and show you the most important approaches and methods for visualizing, smoothing and predicting time series in Python.

After the course you will be able to use time series analysis with Python independently in your daily work.

Table of contents: 

• Visualization and decomposition of time series
• Aggregation and stationary processes of time series
• Calculation of ARIMA and SARIMA models
• Calculation of exponential smoothing models
• Calculation of regressions with SARIMA errors
• Forecasting

—

Recommended course length: One day