Introduction to Data Science 2022/23 fall

Practice sessions

• Attending at least 9 of the 12 practice sessions is compulsory: after missing 3 practice sessions, each additional missed practice session results in losing 5 points. You can check how many sessions you attended - Please log in to check the attendances. (link coming here soon)

Groups 1, 2, 3 and 7 only: mark your attendances - Please log in to mark the attendances.

Before the first practice session

Before the first practice session please make sure that you have Python version 3 and Jupyter Notebook installed on your computer. If you don't have it yet, then we recommend to choose between the following options:

1. Full Anaconda (can take up to 3 Gb of space):
   • Install Anaconda (Jupyter Notebook is included)
     • Anaconda (it has most of the necessary packages installed and GUI, however, it takes 3 Gb of space)
2. Mini-version of Anaconda (takes less space):
   • Install MiniConda and Jupyter Notebook
     • MiniConda (it only installs Python and Conda, no unnecessary space taken, more installing will be needed later)
     • conda install jupyter (install Jupyter Notebook)
3. No Anaconda, relying on the system Python:
   • Installing Jupyter Notebook to the system environment
     • In these instructions see the last section to install with pip and without Anaconda

After installing Jupyter Notebook you can simply run Jupyter notebook from command line:

• jupyter notebook (alternatively open Anaconda prompt to open Jupyter Notebook)

If you use Python in multiple courses and potentially different versions of packages, then you might want to create an environment into which you would install the packages needed for this course:

• Optional: Creating and using an environment:
  • Creating a new virtual environment for this course after installation of Conda
    • conda create -n IDS python=3.9 (create a new environment)
  • Now you should activate the environment before launching Jupyter notebook:
    • conda activate IDS or activate IDS (Windows) or source activate IDS (Linux, Mac) - activates the environment
    • conda install jupyter (make sure the environment is activated before installing)
    • jupyter notebook (make sure the environment is activated)
  • Any further installations must then also be done within the environment
  • The environment can be closed by conda deactivate

Practice session 01 - Sept 5-7 - Introduction

Useful materials: Jupyter Notebook Tutorial video
Practice materials: Introduction to Jupyter Notebook & Introduction to Data Science notebooks

Practice session 02 - Sept 12-14 - DataFrames and Plotting

Notebooks (zip): tutorial for dataframes and plotting

Practice session 03 - Sept 19-21

Extra material: Pandas

Practice session 04 - Sept 26-28

Extra material: NumPy & SciPy tutorial

Practice session 05 - Oct 3-5

Extra material: Feature Engineering

Practice session 06 - Oct 10-12

Extra material: clustering

Practice session 07 - Oct 17-19

Extra material: Grid Search

Practice session 08 - Oct 24-26

Extra material: Sampling

Practice session 09 - Oct 31-Nov 2

Extra material: Linear regression

Practice session 10 - Nov 7-9

Project introduction presentations (information here).

Practice session 11 - Nov 14-16

Extra material: TensorFlow tutorial

Practice session 12 - Nov 21-23

Teams work on projects - Nov 24-Dec 15