Python for Data Science

Overview

This course is designed for engineers, analysts, and technical professionals who want to gain practical data science skills using Python. The training focuses on real-world applications such as production monitoring, quality analysis, and process optimization.

Participants will learn how to work with data using industry-standard Python libraries, perform exploratory analysis, build visualizations, and create machine learning models using real engineering data. The course includes a final mini-project tailored to your company’s domain, allowing immediate transfer of knowledge into practice.

Requirements

Solid understanding of Python (data types, functions, loops, classes).

Certificate

Participants receive a certificate of completion of the authorized Softinery course.

Materials and Tools

You will have access to all the course materials: course notes, exercise solutions and presentations. Moreover, you will learn the basics of most important tools used in modern Python applications in data science.

NumPy

Python for Data Science – Syllabus

What is Data Science? Role in engineering and production
Typical data science pipeline
Overview of tools: Jupyter, pandas, NumPy, matplotlib, scikit-learn
Working with Jupyter Notebooks for reproducible analysis

DataFrames and Series: structure, indexing, slicing
Loading data: CSV, Excel, SQL, JSON
Handling missing data
Filtering, sorting, grouping, aggregation
Joining and merging datasets
Pivot tables and reshaping
Efficient iteration and vectorized operations

Identifying and handling outliers
Scaling and normalizing data
Encoding categorical variables
Feature extraction and transformation
Automating preprocessing pipelines
Working with time series data

Descriptive statistics (mean, median, std, percentiles)
Correlation and covariance
Group-based analysis
Pattern discovery using slicing and grouping
Real-world EDA on industrial/engineering datasets

Supervised vs unsupervised learning
The role of ML in modern process engineering
Data splitting and cross-validation
Building a pipeline for ML
Basic models:
- Linear Regression
- Decision Trees
- KMeans Clustering
Model evaluation (accuracy, confusion matrix, F1-score)

Apply complete pipeline: loading, cleaning, analyzing, modeling
Use real industrial data (can be customized to client’s domain)
Build report or dashboard based on insights and predictions
Present and document results in Jupyter Notebook or PDF

What Our Customers Have To Say

Michael
“Softinery course is a perfect opportunity for students to upgrade their skills. I feel have strong foundation based on which I can progress with the projects that interest me.”

Guest
“The course is great for anyone trying to get started in pandas and numpy. Above all, It was challenging and relevant. Highly recommend this course..”

Ivan
“The right blend of teaching and practicing. Simon is a great teacher and very patient in helping you. I feel confident in my ability to continue my python learning independently.”

Instructor

PhD Eng. Szymon Skoneczny is a former university professor specialized in mathematical modelling. He has also worked for international companies like Electricite de France, Siemens and ArcellorMittal.

Over 3000 hours of tutoring

Specialized in mathematical modelling

Over 40 scientific articles

Experienced in algorithms and high-performance computations

FAQ

The training is aimed at people who know the basics of programming in any language. If you have ever programmed in C/C++, Java, Matlab or other language then you can deal with problems discussed during the training easily. Will need to put some effort for installation of Python and packages. It will be discussed on training briefly. You will also get help if necessary.

Each subject of the course is divided into theoretical and practical part. In average for every 20 minutes of presentation there will be 40 minutes of practice. Because groups are small (12 people) everyone can get help.

The maximum number of participants is 12. Because the training aims at practice it is necessary to keep the group small.