Python training for data science

The introductory Python training course with a focus on data science is your ideal entry point into the programming language.
In addition to programming paradigms, object types, and syntax structures, the central library for data management, polars, is covered.
By the end of the training, you will be able to implement data management as well as initial analyses and visualizations in Python.

Course contents

• Introduction to Python
• Introduction to the Jupyter Notebooks development environment
• The philosophy of Python (the programming language, objects, functions, classes)
• Object types and their properties (list, dictionary, string)
• Data management with polars
• Analysis with Python: statistical measures and charts
• Writing your own functions and loops
• Introduction to object orientation: classes, methods, and attributes

—

Target group: This introductory course is aimed at participants with no prior knowledge of Python or statistics and serves as a starting point for further work with Python.

—

Course duration: Two days

A selection of training contents in the area of machine learning with Python

• Introduction to the basic concepts of machine learning
• Working with the scikit-learn machine learning framework
• Logic and practice: splitting data into training and test sets
• Presentation and application of relevant preprocessing steps, e.g. one-hot encoding, standardization, imputation
• Training machine learning models based on various algorithms, e.g. decision trees, support vector machines, random forests, XGBoost, etc.
• Combining preprocessing and modeling steps in a scikit-learn pipeline
• Interpreting various metrics for model evaluation
  • For classification: (balanced) accuracy, AUC, F1 score, etc.
  • For regression: RMSE, MAE
• Model parameter tuning (grid search) using cross-validation
• Independent application of what has been learned in practical exercises

—

Target group: This introductory course is aimed at participants who already have prior knowledge of Python or statistics.

—

Course duration: Two days

Transformer models have revolutionized the way computers process and understand text. They form the foundation of modern AI applications.
In our training, we cover the background and use cases of transformer models and teach the skills needed for their practical application.

Course contents at a glance:

• Introduction: overview of use cases for transformer models (text, audio, image)
• Background of transformer models
• Practical work with transformers in Huggingface
• What is Huggingface, and what do I need it for?
• How do I process data and prepare it for working with transformers? (datasets library)
• How do I apply transformer models in various use cases? (transformers library)
• How do I evaluate model performance? (evaluate library)
• How can I finetune a pretrained model for my application?

—

Course duration: Two days

Optionally, we can extend the transformer model training by an additional day for a total duration of three days.

Contents of the optional extension day:

• Transformers in production: introduction to a RAG system
  • Improving speed or reducing RAM usage with quantization and the accelerate library
  • The role of transformers in a RAG system
  • How to build a RAG system: langchain library
  • Prompt engineering
  • (Presenting RAG as an app using gradio)