Machine Learning in Practice – Python and Predictive Models

Training process

Training needs analysis

If you have specific requirements regarding the training programme, we will carry out a training needs analysis for you. This will guide us on which aspects of the programme should receive greater emphasis, so that the training programme meets your specific needs.

What will you gain?

End-to-end ML workflow - You will structure the full predictive modeling process, from splitting data and preparing datasets to training, evaluating, and improving models within a practical machine learning pipeline.

Linear regression in Python - You will build a linear regression model in Python yourself, understand its assumptions, and learn how to read the results so you can turn them into useful business insights.

Stronger data preparation - You will learn how to clean data, handle missing values, scale features, and encode categorical variables, so you can prepare inputs that genuinely improve model performance.

Confident model evaluation - You will choose the right metrics for regression and classification, including RMSE, MAE, R², Precision, and AUC, and interpret results correctly instead of guessing model quality.

Overfitting under control - You will learn to spot overfitting and underfitting, use cross-validation, and apply Ridge and Lasso regularization, so your models perform well not only on training data but also on new cases.

Tuning with pipelines - You will practice Grid Search, explore the basics of Random Search, and build scikit-learn pipelines, helping you test model variants faster and keep experiments organized.

Classification methods compared - You will compare logistic regression, decision trees, Random Forest, and KNN, so you can select an algorithm that fits a classification task instead of relying on a single method.

Practical text classification - You will create a basic text classification model using Bag of Words and TF-IDF representations, giving you a solid starting point for sentiment analysis and content categorization work.

Training programme

1. Introduction to machine learning

types of machine learning,
model building process (ML pipeline),
data split (train/test),
tools and libraries (scikit-learn, pandas, numpy).

2. Regression – introduction (linear regression)

what regression is,
linear regression – assumptions and operation,
building a regression model in Python,
interpretation of the model results.

3. Regression – preprocessing and data preparation

data cleaning,
handling missing values,
scaling and normalization,
encoding categorical variables,
Feature engineering (introduction).

4. Regression – evaluation metrics

MSE, RMSE,
MAE,
R² (coefficient of determination),
interpretation of model results.

5. Regression – overfitting and regularization

overfitting vs underfitting,
cross-validation,
regularization (L1, L2 – Ridge, Lasso),
selection of model complexity.

6. Regression – model optimization

introduction to hyperparameter tuning,
Grid Search,
Random Search (introduction),
Pipeline in scikit-learn.

7. Classification – algorithms

introduction to classification,
logistic regression,
decision trees,
Random Forest (introduction),
KNN (k-Nearest Neighbors).

8. Classification – model evaluation

Accuracy, Precision, Recall, F1-score,
confusion matrix,
ROC and AUC,
model validation.

9. Text classification

introduction to NLP in ML,
text representation (Bag of Words, TF-IDF),
building a text classification model,
examples of applications (sentiment analysis, categorization).

10. Good practices for working with ML models

end-to-end pipeline,
reproducibility of results,
interpretability of models,
preparation of models for deployment.

What are the prerequisites for participating in the training?

Basic Python skills - You should be comfortable writing simple Python scripts, using variables, loops, functions, and imports, so you can focus on models rather than learning syntax.

Working with tabular data - You should understand rows, columns, data types, and basic table operations, so you can work efficiently with input data in pandas and prepare datasets.

Math fundamentals - You should know basic algebra and statistics, including averages, variation, and relationships between variables, to better understand models and evaluation metrics.

Development environment basics - You should know how to run a notebook or script, install libraries, and work in a basic coding environment, so you can complete the hands-on exercises smoothly.