Details: Construction, resolution, and interpretation of probabilistic models.
Data analysis and machine learning tools are used. Among the tasks are:
Methodology: Bayesian models are implemented using mainly the Python libraries PyMC and ArviZ. The flexibility of modeling tools allows us to capture the structure of the data, for example through hierarchies, and prior domain knowledge. If necessary, the libraries used can be adapted to allow, among other demands, the scalability of inferences. In addition to probabilistic modeling, there is the ability to apply other forms of modeling including mathematical modeling, computational simulations, and Machine Learning techniques.
Details: Teaching courses related to scientific programming using the Python programming language
and its associated libraries. The training may incorporate the following topics:
Methodology: Applicant-centered, problem-based Python training. For this, tools from the Python ecosystem are used such as Jupyter Notebooks, IDEs, and scientific libraries such as Numpy or Matplotlib, among others.