Dear students,
The first practical session for the course INFO-F422 Statistical Foundations of Machine Learning will be held on Thursday 20/02 from 4:00pm to 6:00pm at Forum C of the Plaine campus. Subsequent sessions will also be on Thursdays from 4:00pm to 6:00pm at Forum C. Please refer to TimeEdit for more details.
The initial session will focus on providing a practical introduction to Python and Monte Carlo simulations for statistical analysis. Python will be utilized throughout lectures, practicals, projects, and exams to implement and illustrate concepts.
We assume you are already familiar with Python. If not, it's essential to start learning Python independently as soon as possible.
Note: it is crucial that we all use the same environment to ensure compatibility and a smooth learning experience. We have decided on Python 3.12 (to avoid known compatibility issues with certain libraries on newer versions and some operating systems).
We strongly recommend creating a conda environment, as it conveniently manages both Python versions and packages in a single step. If you install conda (e.g., through Anaconda or Miniconda), you do not need a separate Python installation.
1. Install Conda (Anaconda or Miniconda) for your operating system.
2. Create a new environment with Python 3.12:
conda create --name sfml python=3.12 numpy scipy pandas scikit-learn xgboost pytorch matplotlib seaborn notebook jupyterlab statsmodels tqdm -c conda-forge
3. Activate the environment:
conda activate sfml
Over the course of the practical sessions, we will install additional libraries (via pip) as needed within this environment.
Note that:
• We will provide support only for this Conda-based environment.
• We will not provide support for other environment managers (e.g., using PyCharm’s built-in environment creation or system-level Python installations) because these often lead to unnecessary compatibility issues.
• Your project and exam will be evaluated using Python 3.12. Any issues arising from version incompatibility will be penalized as a failure to adhere to the guidelines.
Resources to start learning Python:
• Python Documentation - Comprehensive guides and tutorials: https://docs.python.org/3/
• Codecademy - Interactive Python course: https://www.codecademy.com/learn/learn-python-3
• Automate the Boring Stuff with Python - Free online book: https://automatetheboringstuff.com/
• Real Python - Python tutorials and courses: https://realpython.com/
If you encounter significant issues with Python that hinder your course progress, please contact me promptly via email at Cedric.Simar@ulb.be, and I'll assist you as best as I can.