Data Science with Python Seminar
In-person events in Wiesbaden or online seminar on 2 days: €1,190 per person (net)
If you want to get started with data science using Python quickly and efficiently, this seminar is for you!
Open training dates: October 28/29, 2024, March 27/28, 2025
LEARNING OBJECTIVES AND AGENDA
Note: The focus is now on Data Preparation , AutoML and MLflow
Learning objectives :
The four important data science packages are: NumPy, Scikit learn, pandas, Matplotlib.
Data Preparation : Relevant data preparation methods: handling missing values, standardization, outliers, implausible cases. Variable transformations. Handling dummy variables. Handling text, and much more.
- Overview of relevant machine learning methods . A classification or regression case will be analyzed in the seminar.
- Model evaluation and tips for controlling overfitting .
- Automating modeling with AutoML . Hyperparameter tuning with AutoML. Feature selection .
- Mapping the entire process using a pipeline : MLflow in practice.
- Using the two relevant deep learning frameworks : pyTorch and TensorFlow . A model is calculated here.
IN-HOUSE SEMINAR
Seminars held at the customer's location
€1,390.00
per day up to 4 participants plus statutory VAT
All content of the in-house seminars is individually tailored and taught to specific target groups .
Intensive follow-up support enables participants to implement their knowledge in the shortest possible time.
Recommended seminar duration: 2 days
Rental fees for training notebook (on request): 60,- Euro (per day, per training computer)
WORKSHOP
You tell us your topics!
Price on request
plus statutory VAT and travel expenses if applicable
All workshop content is individually tailored and taught to specific target groups .
We are happy to conduct the workshop at your location, in Wiesbaden or online.
Rental fees for training notebook (on request): 60,- Euro (per day, per training computer)
Day 1
Python basics for data science, help, overview of essential resources
Interactive work in data science projects with Jupyter notebooks
Importing data using pandas from data sources such as SQL, MS Excel, etc.
Analyze the data using pandas and matplotlib, as well as other packages. The focus is on implausible or missing values, outliers, distributions, descriptive statistics and graphs, correlations/heatmaps, etc.
Preparing the data for analysis purposes based on the insights gained: replacing missing values, distribution transformations, standardizations, obtaining sub-datasets, splitting into test and training datasets, etc.
Day 2
Modeling algorithms part 1: Basic procedures for supervised learning: Logistic regression analysis and simple trees for categorical target variables as well as multiple linear regression models for metrically scaled target variables (possibly on the first day, depending on the progress of the group)
Model evaluation: Typical outcome parameters according to the measurement level of the target variable (MAE, MAPE, R-squared), accuracy, confusion matrix, AUC/ROC curves, overview of other parameters)
Modeling Algorithms Part 2: Advanced (Black Box) Methods: Random Forests, Gradient Boosting, etc.
Ensemble method: Ensemble method, Bagging (Bootstrap Aggregation)
Hyperparameter tuning: strategies, grid search, random search, model selection
Deep learning with TensorFlow and PyTorch: basics and simple models.
CONTENTS
This two-day compact course teaches the essential skills needed to tackle machine learning problems using Python.
Why Python? TIOBE selected Python as its programming language of the year 2020, particularly because of its use in machine learning. With numerous add-on packages —numpy, pandas, scikit-learn, matplotlib, plotly, etc.—Python is ideal for analyzing and preparing data, applying machine learning models, and deploying the resulting models.
Interactive work : Unlike traditional programming, developing machine learning models is an interactive process. At the beginning of a project, it's rarely clear what the data will yield, what data preparation is necessary, and how complex it will be to train a large number of machine learning models. This interactive process is optimally supported by the use of Jupyter notebooks, as they allow not only Python code but also the insertion of text using Markdown. Therefore, we conduct all exercises in Jupyter notebooks.
Seminar structure : The seminar follows the CRISP-DM model, whose first steps ( Business Understanding, Data Understanding ) offer significant added value for successful implementation in data science projects. The subsequent steps—data preparation and modeling—are often implemented in a highly agile manner today. The remaining two steps: evaluation (with regard to achieving the business objectives) and deployment, are intended to further emphasize that data science projects are not developed in a vacuum, but rather should, for example, deliver a positive return on investment or fulfill other business objectives .
Seminar content : The seminar content has been developed by data science experts , thus addressing your practical business needs . A detailed overview of the seminar content can be found in the agenda . All topics are always implemented practically using the Jupyter notebooks mentioned above. The pace depends on the group's progress. The very small group sizes (in-person events in Wiesbaden are held with a maximum of four participants) allow for individual questions to be addressed.
Offer guarantee : Our service for you: The face-to-face seminars in Wiesbaden will be held with just one registration !