Machine Learning Catalogue
From beginner to expert — buy individually or as a bundle with a certificate of completion.
Machine Learning roles turn business needs and data into reliable predictive models that are evaluated, optimized, deployed and monitored. This learning path develops skills used in particular by Data Scientists and Machine Learning Engineers, with deeper focus on models, their evaluation, optimization and technical lifecycle, from experimentation through production deployment. These Machine Learning skills are then further developed and placed within a broader business-oriented approach in the Data Scientist learning path, which also covers business framing, experimentation, causality, complex data and Data Science product management.
Each course brings together 8 interactive lessons in a Learning Workspace designed to turn knowledge into operational skills. Guided lessons, dynamic quizzes, conversational support, practical cases, additional resources and progress tracking let you learn effectively, at your own pace. The 15 courses follow a level-by-level progression, from framing a Machine Learning problem to deploying, monitoring and defending a complete predictive solution.
🎓 15 courses • 📚 120 interactive lessons • ⏱️ 45 to 75 hours of conversational learning
⭐ Best value
Machine Learning — Complete Pack (15 courses)
ML01 → ML15 · 1 800 credits included
142,90 € excl. VAT
Savings : 104,60 €
29,70 €
22,90 excl. VAT
Savings : 6,80 €
Understand Machine Learning and scope a problem
turn a business need into a measurable Machine Learning problem and decide whether ML is really the right approach
Course information
Data Analysts, Data Scientists, Machine Learning Engineers and developers who want to build reliable and reproducible machine-learning solutions.
No specific prerequisite to start. Confidence with digital tools is recommended; later technical courses may use development or data concepts.
- Understand what Machine Learning really brings
- Identify classification, regression and other problem families
- Build a baseline before any complex model
- Workshop: decide whether a problem truly requires Machine Learning
- Turn a business need into a measurable Machine Learning problem and decide whether ML is really the right approach
- Distinguish supervised and unsupervised learning
- Distinguish training, validation and inference
- Workshop: decide whether a problem truly requires Machine Learning
3 to 5 h
9,90 € excl. VAT
Prepare and explore data for Machine Learning
build a usable dataset without introducing bias, data leakage or inconsistencies
Course information
Data Analysts, Data Scientists, Machine Learning Engineers and developers who want to build reliable and reproducible machine-learning solutions.
A working knowledge of the foundations covered earlier in the pathway is recommended. This course builds on concepts introduced in previous courses.
- Explore distributions, types and data quality
- Prepare numerical and categorical variables
- Spot imbalances and insufficient representativeness
- Workshop: audit and prepare a dataset before modeling
- Build a usable dataset without introducing bias, data leakage or inconsistencies
- Handle missing values and outliers
- Scale variables when the model requires it
- Workshop: audit and prepare a dataset before modeling
3 to 5 h
9,90 € excl. VAT
Build a first reproducible ML pipeline
chain preprocessing, training, prediction and evaluation into a reproducible workflow
Course information
Data Analysts, Data Scientists, Machine Learning Engineers and developers who want to build reliable and reproducible machine-learning solutions.
A working knowledge of the foundations covered earlier in the pathway is recommended. This course builds on concepts introduced in previous courses.
- Set up a reproducible ML environment
- Create preprocessing transformations
- Calculate and verify first metrics
- Workshop: deliver a first reproducible ML pipeline
- Chain preprocessing, training, prediction and evaluation into a reproducible workflow
- Build a baseline with scikit-learn
- Train the pipeline and generate predictions
- Workshop: deliver a first reproducible ML pipeline
3 to 5 h
9,90 € excl. VAT
38,70 €
28,90 excl. VAT
Savings : 9,80 €
Solve regression problems
design, compare and evaluate models intended to predict a numerical value
Course information
Data Analysts, Data Scientists, Machine Learning Engineers and developers who want to build reliable and reproducible machine-learning solutions.
A working knowledge of the foundations covered earlier in the pathway is recommended. This course builds on concepts introduced in previous courses.
- Understand a continuous target and build a baseline
- Understand regularization and its benefits
- Interpret R² without overestimating it
- Workshop: compare several regression models on a business need
- Design, compare and evaluate models intended to predict a numerical value
- Build a linear regression
- Choose between MAE, MSE and RMSE
- Workshop: compare several regression models on a business need
3 to 5 h
12,90 € excl. VAT
Solve classification problems
build classification models and choose metrics and thresholds suited to the real cost of errors
Course information
Data Analysts, Data Scientists, Machine Learning Engineers and developers who want to build reliable and reproducible machine-learning solutions.
A working knowledge of the foundations covered earlier in the pathway is recommended. This course builds on concepts introduced in previous courses.
- Understand classes and build a baseline
- Use trees and Random Forest
- Compare precision, recall and F1
- Workshop: choose a classifier based on the cost of errors
- Build classification models and choose metrics and thresholds suited to the real cost of errors
- Build a logistic regression
- Read a confusion matrix
- Workshop: choose a classifier based on the cost of errors
3 to 5 h
12,90 € excl. VAT
Evaluate correctly and compare models
get a reliable performance estimate and avoid misleading conclusions when selecting a model
Course information
Data Analysts, Data Scientists, Machine Learning Engineers and developers who want to build reliable and reproducible machine-learning solutions.
A working knowledge of the foundations covered earlier in the pathway is recommended. This course builds on concepts introduced in previous courses.
- Understand the risk of an overly optimistic evaluation
- Evaluate with cross-validation
- Compare several models on consistent metrics
- Workshop: build a defensible evaluation protocol
- Get a reliable performance estimate and avoid misleading conclusions when selecting a model
- Use a hold-out set correctly
- Respect temporal order with a time split
- Workshop: build a defensible evaluation protocol
3 to 5 h
12,90 € excl. VAT
47,70 €
35,90 excl. VAT
Savings : 11,80 €
Design and select features
create informative variables without data leakage and reduce complexity when it improves the model
Course information
Data Analysts, Data Scientists, Machine Learning Engineers and developers who want to build reliable and reproducible machine-learning solutions.
A working knowledge of the foundations covered earlier in the pathway is recommended. This course builds on concepts introduced in previous courses.
- Understand the role of feature engineering
- Create time-based features without information leakage
- Use feature selection methods
- Workshop: improve a model with justified features
- Create informative variables without data leakage and reduce complexity when it improves the model
- Create useful numerical variables and transformations
- Detect redundancy and multicollinearity
- Workshop: improve a model with justified features
3 to 5 h
15,90 € excl. VAT
Optimize models and hyperparameters
reduce underfitting and overfitting and search hyperparameters without contaminating the final evaluation
Course information
Data Analysts, Data Scientists, Machine Learning Engineers and developers who want to build reliable and reproducible machine-learning solutions.
A working knowledge of the foundations covered earlier in the pathway is recommended. This course builds on concepts introduced in previous courses.
- Understand underfitting, overfitting, bias and variance
- Build a Grid Search with cross-validation
- Control compute time and search space
- Workshop: conduct and document a hyperparameter optimization
- Reduce underfitting and overfitting and search hyperparameters without contaminating the final evaluation
- Identify learned parameters and hyperparameters
- Choose the metric actually being optimized
- Workshop: conduct and document a hyperparameter optimization
3 to 5 h
15,90 € excl. VAT
Explore unlabeled data
use clustering, dimensionality reduction and anomaly detection to understand structures without a supervised target
Course information
Data Analysts, Data Scientists, Machine Learning Engineers and developers who want to build reliable and reproducible machine-learning solutions.
A working knowledge of the foundations covered earlier in the pathway is recommended. This course builds on concepts introduced in previous courses.
- Understand the goals of unsupervised learning
- Segment with K-means
- Reduce dimensionality with PCA
- Workshop: build and interpret an exploratory segmentation
- Use clustering, dimensionality reduction and anomaly detection to understand structures without a supervised target
- Prepare data for clustering
- Discover DBSCAN and irregularly shaped clusters
- Workshop: build and interpret an exploratory segmentation
3 to 5 h
15,90 € excl. VAT
56,70 €
42,90 excl. VAT
Savings : 13,80 €
Interpret, explain and challenge models
understand the factors that influence predictions and identify situations where an explanation can be misleading
Course information
Data Analysts, Data Scientists, Machine Learning Engineers and developers who want to build reliable and reproducible machine-learning solutions.
A working knowledge of the foundations covered earlier in the pathway is recommended. This course builds on concepts introduced in previous courses.
- Distinguish predictive performance and explainability
- Analyze feature importance with caution
- Discover SHAP-style local explanations
- Workshop: produce an explainability analysis with its limits
- Understand the factors that influence predictions and identify situations where an explanation can be misleading
- Interpret coefficients and simple parameters
- Understand partial dependence plots
- Workshop: produce an explainability analysis with its limits
3 to 5 h
18,90 € excl. VAT
Manage imbalance, bias, robustness and reliability
assess a model's quality beyond an average metric and control its risks on populations and hard cases
Course information
Data Analysts, Data Scientists, Machine Learning Engineers and developers who want to build reliable and reproducible machine-learning solutions.
A working knowledge of the foundations covered earlier in the pathway is recommended. This course builds on concepts introduced in previous courses.
- Identify the consequences of strong class imbalance
- Choose metrics and thresholds for rare cases
- Test robustness and sensitivity to perturbed data
- Workshop: challenge a model's reliability before a decision
- Assess a model's quality beyond an average metric and control its risks on populations and hard cases
- Use weighting and resampling methodically
- Compare performance across subgroups
- Workshop: challenge a model's reliability before a decision
3 to 5 h
18,90 € excl. VAT
Track experiments and manage model versions
make experiments comparable, auditable and reproducible, then manage the model lifecycle
Course information
Data Analysts, Data Scientists, Machine Learning Engineers and developers who want to build reliable and reproducible machine-learning solutions.
A working knowledge of the foundations covered earlier in the pathway is recommended. This course builds on concepts introduced in previous courses.
- Understand experiment, run and ML artifact
- Track models, figures and artifacts
- Manage versions, aliases, tags and metadata
- Workshop: build the full traceability of a candidate model
- Make experiments comparable, auditable and reproducible, then manage the model lifecycle
- Log parameters and metrics with MLflow Tracking
- Register a model in the Model Registry
- Workshop: build the full traceability of a candidate model
3 to 5 h
18,90 € excl. VAT
74,70 €
48,90 excl. VAT
Savings : 25,80 €
Understand and practice Deep Learning
understand the essential mechanisms of neural networks and train a simple model without diluting the path in advanced Deep Learning
Course information
Data Analysts, Data Scientists, Machine Learning Engineers and developers who want to build reliable and reproducible machine-learning solutions.
A working knowledge of the foundations covered earlier in the pathway is recommended. This course builds on concepts introduced in previous courses.
- Understand tensors, layers and neural networks
- Understand the loss function and optimization
- Build and train a simple network with PyTorch
- Workshop: compare a simple network to a classic baseline
- Understand the essential mechanisms of neural networks and train a simple model without diluting the path in advanced Deep Learning
- Choose activation functions and a simple architecture
- Use batch, epoch and DataLoader
- Workshop: compare a simple network to a classic baseline
3 to 5 h
24,90 € excl. VAT
Deploy, monitor and maintain an ML model
move from a validated model to a usable, versioned, observable and maintainable production service
Course information
Data Analysts, Data Scientists, Machine Learning Engineers and developers who want to build reliable and reproducible machine-learning solutions.
A working knowledge of the foundations covered earlier in the pathway is recommended. This course builds on concepts introduced in previous courses.
- Prepare a reliable inference artifact
- Expose a model through an inference API
- Monitor latency, errors and input data quality
- Workshop: deploy and monitor a prediction service
- Move from a validated model to a usable, versioned, observable and maintainable production service
- Distinguish batch prediction and online inference
- Manage registry, versions, promotion and rollback
- Workshop: deploy and monitor a prediction service
3 to 5 h
24,90 € excl. VAT
Full professional Machine Learning project
design, evaluate, explain, version, deploy and defend an end-to-end ML solution
Course information
Data Analysts, Data Scientists, Machine Learning Engineers and developers who want to build reliable and reproducible machine-learning solutions.
A working knowledge of the foundations covered earlier in the pathway is recommended. This course builds on concepts introduced in previous courses.
- Scope the churn problem and the cost of errors
- Build pipelines, features and candidate models
- Track experiments and version the final model
- Final workshop: present and defend the ML solution and its limits
- Design, evaluate, explain, version, deploy and defend an end-to-end ML solution
- Prepare data and build the baseline
- Analyze segments, errors and explainability
- Final workshop: present and defend the ML solution and its limits
3 to 5 h
24,90 € excl. VAT