Python for data analysis - You will learn to work confidently with lists, dictionaries, and NumPy arrays, so you can prepare data faster for analysis, reporting, and machine learning tasks used in everyday work.
Practical business scripts - You will see how to build functions and classes that support ML projects, then use them to automate sales data analysis and streamline repetitive tasks in a practical business context.
Choosing the right algorithm - You will understand when to use supervised, unsupervised, or other learning approaches, so you can match methods more accurately to customer behavior prediction and business goals.
Hands-on scikit-learn - You will practice building models with scikit-learn from data preparation through training and testing, so you can run a complete machine learning workflow in Python on your own.
Core data science libraries - You will gain practical skill in NumPy, Pandas, Matplotlib, and SciPy, making it easier to combine data wrangling, statistical analysis, and visualization in one workflow.
Model quality evaluation - You will learn model evaluation methods and testing good practices, so you can tell whether a model truly performs well instead of relying on results that only look convincing.
Regression and tree models - You will learn how to use linear and polynomial regression as well as decision trees and Random Forest to predict numeric values and compare the effectiveness of different model types.
Clear analysis and visuals - You will create charts and examine data correlations, helping you understand patterns in datasets and present model outcomes in a way that is clear for both your team and business users.
Python for Data Science begins with structures that support everyday analytical work. Participants use lists, dictionaries, and NumPy arrays, then connect functional and object-oriented programming with business data processing in a sales analysis automation project.
Supervised, unsupervised, and reinforcement learning define the main directions of machine learning work. The section connects these approaches with popular algorithms, business use cases, practical scikit-learn usage, and a project on predicting customer behavior.
At the organizational level, it is important to separate broader AI concepts from the narrower scope of Machine Learning. The section clarifies this relationship and places ML within the wider landscape of intelligent systems and data-driven automation.
NumPy, Pandas, Matplotlib, and SciPy form the technical base for many Machine Learning workflows in Python. Their roles are discussed in relation to numerical operations, tabular data handling, visualization, and scientific computing used throughout analytical tasks.
Choosing a model cannot rely on intuition alone. This section presents methods for evaluating Machine Learning models, with attention to how performance is measured, compared, and interpreted so that results can support a justified selection process.
When the goal is to estimate a number rather than assign a class, regression becomes the central approach. The material covers linear and polynomial models, showing how they are used to predict numerical values and where each form fits the data structure.
Building predictive rules from data requires understanding how tree-based methods split observations. Here the core algorithms behind Decision Trees are explained, along with their theoretical assumptions and the differences that distinguish them from linear methods.
Random Forest combines multiple decision trees to improve stability and prediction quality. The section explains the method’s theoretical basis, reviews the main parameters, and shows how parameter handling affects model behavior, complexity, and final results.
Charts support the move from raw datasets to interpretable conclusions. Participants work with key chart types and learn how correlations between variables are identified, compared, and presented so analytical findings can guide later modeling choices.
Reliable deployment starts with disciplined model testing. This part focuses on verifying Machine Learning models with repeatable checks, separating training results from real performance, and spotting issues that can distort conclusions before a model is used in practice.
Basic computer skills - You should be comfortable using a computer, installing software, managing files, and launching applications so you can complete the hands-on exercises without difficulty.
Basic logic and math - You should understand simple algebra, the idea of a variable, percentages, and relationships between values so you can analyze data more easily and interpret model results.
Experience with data tables - You should have basic experience working with data tables, for example in Excel or a similar tool, so you can understand records, columns, and simple data operations quickly.
Readiness to write code - You should be ready to write and revise simple code on your own, because the training includes practical exercises and active work with examples in Python.
