Data Science Course
Full Time
(9 Weeks)
Part Time
(24 Weeks)
Flexible
Payment
About The Data Science Course
Learn data science from Python to advanced Machine Learning, get all the skills to join a data science team and boost your career. Le Wagon is an immersive experience that takes you from beginner to junior data scientist within weeks. At the end of the course, you will learn to explore, clean and transform data into actionable insights and how to implement Machine Learning models from start to finish in a production environment, working in teams with the best-in-class tool belt.
Our data science course in Mauritius is designed to make you explore, clean, analyse and predict data can lead to different paths The course can be done full time (9 Weeks) and Part-time (24 Weeks).
After the bootcamp, our students are granted lifetime access to our online platform with up-to-date videos and tutorials about the latest tools and best practices of software development. They also become members of our highly engaged community and network of international talents and teachers who keep helping each other and sharing opportunities on a daily basis.
- Once you graduate from Le Wagon, you belong to a global tech community.
- You also benefit from our career services and we help you connect with the best recruiters looking for talent in data-related roles through networking events, job fairs and coaching.
What you will learn ?
- Master programming applied to Data Science in Python
- Design relational databases and build advanced queries with SQL
- Deploy Machine Learning models in production with Google Cloud Platform
- Build and train fully connected deep neural networks to solve classification and regression problems
- Master programming applied to Data Science in Python
- Grasp the mathematic concepts behind data science: statistics, probability and linear algebra
- Conduct advanced analysis with Jupyter notebook, Pandas and Statsmodels
- Implement Machine Learning supervised and unsupervised models with scikit-learn
- Learn Machine Learning best practices (preprocessing, training and testing, performance metrics, etc.)
Course Outcomes & Opportunities
- Become a data analyst, data engineer, data scientist, or data manager for some of the world’s best tech companies (Google, Uber, Getaround, Trainline, Doctolib, etc…).
- Kickstart a career as freelance data scientist helping companies solve their problems with data from wherever you want, and find freelance opportunities through Le Wagon’s network and 43 worldwide campuses.
- Launch a data service or product as an entrepreneur, found your company, raise money to accelerate your growth and manage your data science team.
Course Structure
- Python prerequisites
- Mathematics prerequisites
- Testing your skills
Our Data Science course requires a basic level of Python & Mathematics. As we want all of our students to succeed, you’ll be able to test your level and refresh your skills before the bootcamp starts.
This module covers the fundamentals of Python and Mathematics for data science.
You’ll learn the basics of programming in Python, how to work with Jupyter Notebook & Jupyter Lab, and will become familiar with powerful Python libraries used in data science, such as Pandas and NumPy, to explore big data sets and conduct statistical analyses.
Additionally, we’ll teach you how to collect data from various sources, including CSV files, SQL queries on relational databases, Google Big Query, APIs and Web scraping. You’ll also learn how to build visualisations in order to transform your data into actionable insights. Finally, you’ll understand the concepts of probability, statistics, and linear algebra that underly Data Analysis and Machine Learning.
Contents
- Python For Data Science
- Relational Database & SQL
- Data Visualization
- Statistics, Probability, Linear Algebra
Skilled Learned
- Using Jupyter Notebook
- Loading and exploring a dataset
- Extracting data from different sources
- Database schema architecture
- Translate a business question into a SQL query
- Advanced manipulations of SELECT
- SQL client software like DBeaver or Metabase
- Turn your data into insights with data visualizations
- Building meaningful charts
- Statistics (random variable, distribution, variance, etc.)
- Probability (central limit theorem, Bayes’s theorem)
- Linear Algebra (matrix calculus, derivatives)
In your first one-week-long mini-project, you’ll learn how to use statistical tools and multivariate regression analysis to answer a real business question like a real data analyst.
You’ll learn how to structure a Python repository with object-oriented programming in order to clean your code and make it re-usable, how to survive the data preparation phase of a vast dataset, and how to find and interpret meaningful statistical results based on multivariate regression models.
Contents
-
Statistical Inferences
Skilled Learned
- Structure a Python project’s folder
- Data Analysis
- Hypothesis (A/B) Testing
- Statistical tools (statsmodels)
- Multivariate regression analysis
In this module, you’ll understand the different classes of machine learning models and their applications. You’ll dive deep into the most used library in Machine Learning: scikit-learn. You’ll start with supervised learning and classic methods like linear and logistic regressions to solve prediction tasks. You’ll then move to unsupervised learning and implement methods like PCA for dimensionality reduction or clustering for discovering groups in a data set. Additionally, we’ll teach you how to identify overfitting and the different techniques available to avoid it. Finally, you’ll learn how to tune and evaluate different models to achieve the best performance using methods like cross-validation and hyperparameter tuning. Along the way, you’ll implement all the essential learning algorithms such as KNN, Support Vector Machines and Ensemble Methods like Random Forests or Gradient Boosting.
- Supervised Learning with the Scikit-learn library
- Preprocessing techniques (vectorization, features selection)
- Train linear and non-linear basic models
- Performance evaluation techniques (cross-validation, holdout)
- Adapt the performance metrics to a given task
- Understand the underlying mechanism hidden behind a model training
- Generalization of models (avoid overfitting via loss-function regularization)
- Hyper-parameter tuning (grid-search etc.)
- Assess your business impact effectively
- Improve your model’s performance
- Unsupervised Learning with the Scikit-learn library
- Problem simplification with Dimensional Reduction
- Clustering models (K-means, etc.)
- Train powerful models with ensemble methods
- Analyse and forecast time series with linear model
- Convert the Machine Learning workflow into an iterable Pipeline
This module is based on real life problems that will challenge you to optimize your features and architecture in order to get the best performance.
Contents
- Neural Networks
- Computer Vision
- Times-Series & Text data
- Deep Learning Made Easy
- Understand the architecture of neural networks
- Build your own networks
- Build and train Convolutional Neural Networks (CNN)
- Image preprocessing and batch data loading
- Data Augmentation
- Transfer Learning (VGG16, etc.)
- Build and train Recurrent Neural Networks (RNN, LSTM, etc.)
- Multiple-output time-series forecasts
- Word Embedding
- Sentiment analysis
- Dense Neural Network architecture
- Performance evaluation and overfitting
- Build and train NN with the Tensorflow Keras library
- Launch online training on GPUs with Google Colab
In one week, learn all the best practices around tackling an exciting ML problem too big to solve on your computer alone, and make its prediction available to the world via an API!
First, we’ll teach you how to become more productive in building a machine learning model by using the right workflow. We will then leverage a library called MLflow to log your multiple experimentations, iteration and tuning. Third, we’ll show you how to train at scale using cloud computing power with Google Cloud AI Platform. Finally, you’ll learn Docker to deploy your code and model to production and make it available to the world using Cloud Run or Kubernetes Engine.
Contents
- Machine Learning Pipeline
- Machine Learning workflow with MLflow
- Train at scale with Google Cloud Platform
Skilled Learned
- From Jupyter Notebook to packaged code
- Setting up a good ML project (folders, files, etc..)
- Continuous Integration (GitHub Actions)
- MLflow for logging model performances
- Using Sklearn-Pipeline (Encoders, Transformers)
- Build your own API
- Containerize with Docker and deploy your API on Cloud Run
- Build and deploy a Web-based front-end app
- Discover the cloud computing with Google Cloud Platform (CGP)
- Use cloud-based storage (Cloud Storage)
- Use virtual machines to train models with GPUs (Google AI Platform, AI Notebooks, Colab)
The goal of this module is to bring together all the components you’ve learned so far and work on real open-ended problems in teams.
Skilled Learned
- Collaborate efficiently in teams of 3-4 people on a real data science project
- Through a common Python repository.
- Use a mix of your own datasets (if you have any from your company / non-profit organisation) and open-data repositories (Government initiatives, Kaggle, etc.).
- Put into practice all the tools, techniques, and methodologies covered in the
- Data Science Course.
- Invent, pitch, design, code, and deploy your fully functioning Data Project, to be used in real-life scenarios and as a portfolio-ready application to showcase your new data science skillset.
Learning Outcomes
- analyse rich data sets
- make complex decisions using data
- discover the power of AI & Machine learning
Career Week
At the end of the bootcamp, you are welcome to join our Career Week. This week gives you the tools you need to take the next steps in your career, whether it is finding your first job in tech, building a freelance career, or launching a start-up.
Workshops
Benefit from a combination of panel discussions, workshops, presentations, and assignments to help you find the right career path.
Becoming job-ready
Prepare your personal profile, complete job applications, prepare for technical challenges, and make a game plan for after the bootcamp!
Inspiration
