Cursus Data scientist

Description of the training

intensive training to acquire the basic skills of data science, covering data collection and cleaning, visualisation, exploratory analysis, as well as machine learning and adapted tools.

Objectives of training

At the end of this training, participants will be able to:

• Block 1: Databases of Data Science - Understanding the basics of data science and the Python ecosystem. Apply statistical concepts and start in supervised machine learning.
• Block 2: Advanced Machine Learning Techniques - Mastering unsupervised algorithms and processing textual and temporal data. Optimize models and introduce Deep Learning.
• Block 3: Advanced projects and industrialization - Apply Deep Learning to computer vision and NLP. Deploy models and supervise them in production.
• Block 4: Final project and professional preparation - Design and implement a complete project. Validate skills and prepare for professional certifications.

Who is this training for?

The training is aimed at a wide audience, including:

The training is aimed at a wide audience, including:

• Developers and software engineers wishing to specialize in data science.
• Data analysts who want to strengthen their machine learning skills.
• Data engineers interested in managing and deploying data pipelines.
• Technical project leaders wishing to understand data science to better manage AI projects.
• Researchers and students in retraining or preparing professional certifications.

Prerequisites

No specific prerequisites are required.

Training programme

Block 1: Databases of Data Science (5 days)

• Objective: To understand the fundamentals of data science and the Python ecosystem.

Day 1 (Matin)

• Content: Introduction to data science, main tools (Python, Jupyter), data manipulation with pandas.

Day 1 (afternoon)

• Objective: To explore data structures and types.
• Content: Exploratory analysis, missing data management, visualization with Matplotlib and Seaborn.

Days 2-3: Statistics and Probabilities Applied to Data Science

• Purpose: Apply mathematical bases to data analysis.
• Content: Descriptive statistics, hypothesis tests, distributions, introduction to probabilities.

Days 4-5: Introduction to Machine Learning

• Purpose: Discover the techniques of supervised machine learning.
• Content: Linear regression, logistic regression, model evaluation (metric, overlearning).

Block 2: Advanced Machine Learning Techniques (5 days) Day 6: Machine Learning Unsupervised

• Objective: To master unsupervised algorithms.
• Content: Clustering (k-means, DBSCAN), reduced dimensionality (ACP, T-SNE).

Day 7: Processing of Time and Text Data

• Objective: Manipulate time series and NLP-specific data.
• Content: Time series with pandas, introduction to NLP with spaCy and NLTK.

Days 8-9: Training and optimization of advanced models

• Objective: Optimize model performance.
• Contents: General methods (random forests, gradient boosting), hyperparameter tuning with GridSearchCV.

Day 10: Introduction to Deep Learning

• Objective: To understand the basics of neural networks.
• Content: Perceptron, backpropagation, setting up a model with TensorFlow.

Block 3: Advanced projects and industrialization (5 days) Day 11: Computer Vision

• Purpose: Apply deep learning to computer vision.
• Content: Convolutive networks, applications to image classification.

Day 12: Natural Language Treatment (NLP)

• Objective: Use NLP tools.
• Content: Advanced techniques: Embeddings, transformations (BERT).

Days 13-14: Deployment and Production

• Objective: Implement a deployment pipeline.
• Content: API with Flask, cloud integration (AWS Sagemaker, Azure ML).

Day 15: Post-deployment monitoring and optimization

• Objective: Supervise models in production.
• Content: Introduction to MLflow, data drift management.

Block 4: Final project and professional preparation (3 days) Day 16: Launch of the final project

• Purpose: Design an end-to-end project.
• Content: Problem definition, data exploration, installation of the processing pipeline.

Day 17: Model Development and Evaluation

• Objective: Implement an efficient solution.
• Content: Model training, tests and adjustments.

Day 18 : Presentation and certification

• Objective: Validate skills.
• Content: Project support, preparation for certifications (AWS, Azure, GCP).

Training assets

• Pedagogical and modular approach: Alternative between theory and practice for better assimilation of concepts.
• Cloud Integration: Strong focus on cloud and distributed solutions.
• Qualified speakers: Specialist trainers with practical experience in the field.
• Educational tools and materials: Access to online resources, live demonstrations and real-life case studies.
• Accessibility: Training is open to all, without advanced technical prerequisites.
• Implementation: Complete project from the end of the modules to consolidate the achievements.
• Preparation for Industry: Focus on standard certifications and tools used in the professional environment.

Pedagogical methods and tools used

• Live demonstrations with data science services.
• Practical workshops and real case studies in various sectors (industry, trade, health).
• Feedback: Sharing best practices and common mistakes in business.
• Simulations and tools: Using simulators for interactive workshops.

Evaluation

• End of training QCM to test the understanding of the concepts addressed.
• Practical case studies or group discussions to apply the knowledge gained.
• Ongoing evaluation during practical sessions.
• Implementation: Complete project from the end of the modules to consolidate the achievements.

Normative References

• Well-Architected Cloud Framework.
• GDPR (General Data Protection Regulation).
• ISO 27001, SOC 2 (Service Organization Control).
• NIST Cybersecurity Framework.