Data Science with Python

In the course "Data Science mit Python", you learn in three days how to use Python for data science projects. You are taught the basic functions of machine learning algorithms and practice how to implement them with Python. You don't need any prior programming knowledge for this.

It starts with your Python development environment: You take your first steps with Python and set up your working environment. Then comes NumPy with lists, arrays and the appropriate data types, plus calculations in arrays, broadcasting and fancy indexing. With Pandas you practice data preparation as well as indexing and slicing of a DataFrame.

Then you read data from different formats and convert them. You learn the central Pandas objects: Series, DataFrame and Index. In addition, there are Universal Functions, Hierarchical Indexing as well as merging and grouping of datasets. Based on this, you build basic descriptive statistics and contingency tables.

For data visualization you work with Matplotlib. You create histograms and adjust legends, colors and annotations, again in independent exercises on the computer.

In the machine learning part, it's about feature engineering and a technical overview of supervised and unsupervised algorithms. In supervised learning you practice regression, decision trees, random forests and support vector machines. In unsupervised learning, K-Means clustering, principal component analysis and Gaussian mixture models are on the agenda. You also learn how to check the generalization of your solution. The algorithms are implemented in the frameworks scikit-learn and TensorFlow.

The seminar is aimed at specialists and managers as well as project leaders for data science projects. Theory and practice alternate, and you exchange ideas directly with experienced trainers and other participants. You book the course as a virtual classroom with a guaranteed date or as an in-person date in Hamburg, Munich, Cologne or Nuremberg.

After participation you receive a certificate of attendance. Additionally, you receive a digital open badge as proof of competence. You retrieve it in your personal customer account and share it on request in social networks.

• Python development environment: first steps with Python and setting up the working environment
• Introduction to NumPy: lists, arrays, and NumPy data types
• Calculations in arrays, broadcasting, and fancy indexing
• Introduction to Pandas: data preparation as well as indexing and slicing a DataFrame
• Reading data from different data formats and converting
• Objects in Pandas: Series Object, DataFrame Object, Index Object
• Universal Function (ufunc) with Pandas and Hierarchical Indexing
• Working with datasets: Merge, Join and Aggregation (Grouping)
• Basic descriptive statistics and contingency tables with Pandas
• Data visualization with Matplotlib: histograms, legends, colors, and annotations
• Feature engineering and overview of Machine-Learning algorithms (supervised, unsupervised)
• Supervised Learning: regression, Decision Trees and Random Forests, Support Vector Machines
• Unsupervised Learning: K-Means clustering, Principal Component Analysis, Gaussian Mixture Models
• Validation: determining the generalization of your own solution
• Practical examples with scikit-learn and TensorFlow for supervised and unsupervised learning