Skip to content

Latest commit

 

History

History
53 lines (36 loc) · 2.75 KB

File metadata and controls

53 lines (36 loc) · 2.75 KB

banner_event

Machine learning fundamentals and hands-on

The course "Machine learning fundamentals", is part of the ICTP-International Workshop on Machine Learning for Space Weather: Fundamentals, Tools and Future Prospects | (smr 3750) Buenos Aires, Argentina, 2022. (https://indico.ictp.it/event/9840/)

Description of the course

The course was developed specially for the scientific community with the objective of showing different concepts and techniques of Machine Learning an these application on some scenarios of Space Weather.

Which are the topics of the course?

The course is divided in two parts: the lectures and the hands-on sessions.

In this course you'll to learn about:

  • Data Science pipeline used in general in the projects
  • Machine Learning (ML)
    • What is Machine Learning?
    • What is a ML-based modeling?
    • Why we can apply ML in SWx?
    • Basic concepts of ML
      • Hiperparameters of a ML techniques (activation funciton, loss function, epochs, batch size, etc)
      • Loss function (in classification and forecasting)
      • Overfiting, underfiting, ideal fit!
      • How to messure the performance of the models? (Metrics)
    • Concepts of techniques for classification, forecasting and clustering
      • Artificial neural networks (ANN)
      • Deep learning (DL)

How to clone and use the repo?

For clone the project in your local PC, you have some options:

  • Option 1: Scroll to the top of this page and make click on the green button "Code". Next, click in "Download ZIP". After this, you can extract the project in your local PC.

  • Option 2: From the terminal of your system, run the next command:

git clone https://github.com/Laboratorio-Computacion-Cientifica/International-Workshop-on-Machine-Learning-for-Space-Weather-Fundamentals-Tools-and-Future-Prospects.git

Python tutorials and hands-on resources

Classes of scientific programming using Python: https://github.com/wtpc / https://github.com/orgs/wtpc/repositories The resources include basic Python + source code management (GIT) + Object Oriented Programming (in Python) + multi-language programming (compiled languages with Python) + debugging and profiling + documentation

Python basics:

If you need to refresh your python skills we recommend to complete the following exercises: https://github.com/wtpc/HO-python