General Information
| Language | English |
|---|---|
| Cost | Free of charge |
| Level | Beginner |
Motivation
The potential applications of machine learning (ML) and Deep learning methods to natural science research are numerous, and include detector development, data analysis techniques and even physical system modeling. These techniques will be essential to ensure the highest-quality and impact of scientific results from existing and future experiments or simulations.
Registration
Due to limited capacities, registrations are limited to 30 seats at this point. Please register here! This main registration will be available on a first come, first serve basis.
There is also a waiting list. In case the main registration is filled, please use this waiting list to express your wish to come. Please keep the dates of the course free if you register for the waiting list. We might inform you on short notice. The waiting list will also serve us further to either organize more teaching capacity for this course, or plan for future follow-up courses.
Workshop Content
This introductory course, held virtually via Zoom, will equip students and young researchers from the community with the basic tools necessary to begin implementing and using ML methods using their data. The curriculum will involve some theoretical discussions but focuses on practical examples of the relevant topics. After this course, learners are able to embark in classification as well as regression projects using scikit-learn or keras.
Prerequisites
If you would like to enroll in this workshop, we expect you to bring along knowledge of how to use python language. The course will be highly interactive and will require you to have:
- a microphone
- at best a headset
- recommended: two screens (one connected to the zoom session or to watch the videos, one for coding along)
- a working python environment running on your laptop
We expect participants to come with knowledge on how to use the python programming language. In particular, we expect that you know how to write functions in python, how to write for loops, how to handle lists, access numpy arrays and make plots of data with matplotlib. Prior knowledge of using a data science library like pandas is beneficial, but not required.
Technical Setup
We recommend you to use a local jupyter installation, at best based on anaconda. Details on how to set this up can be obtained from these setup instructions.
The course will rely on 4 main python libraries that may come preinstalled with your python installation: scikit-learn, seaborn, pandas and tensorflow (offering a keras interface). if you do not find them on your system, please install them before the workshop. If you need help installing them, please consult these instructions.
Am I ready?
If you'd like to prepare for the workshop, please start a jupyter notebook and run a cell with the following commands:
import sklearn
print('sklearn version: ', sklearn.__version__)
import seaborn
print('seaborn version: ', seaborn.__version__)
import pandas
print('pandas version: ', pandas.__version__)
from tensorflow import keras
print('Keras version: ', keras.__version__)
We will ask all registered users to send us the output of the code above as an acceptance test before the workshop.
Other software needed
The videos of the lessons are shared as video files (not on a video playing platform like youtube or vimeo). Please install a video player before the workshop that can play .mkv files. One good open-source choice is vlc.
The videos for the lessons are run on google colab. If you struggle with the installation of anaconda (to use jupyter and the libraries mentioned above), please consider using google colab as a fallback.
Interactive pad
