Data Science Training

Table of contents

1. Installation of required tools
2. Other information
3. Project structure
4. Installation guide for Windows

Installation of required tools and Python environment

1. Installing Miniconda

If you already have Anaconda installed, you can skip this step.

You can download Miniconda from here. Make sure you download the 64-bit version and Python 3.x. This is very similar to Anaconda but more lightweight, it will be used to create and manage the Python environment used.

The Python environment is a specific installation of Python and libraries that you use for a project, you should use different environments for different projects, making sure you do not mix depedencies.

2. Creating conda environment (in command line)

To create the Python environment to be used, follow the following steps:

conda create --name ds python=3.7
conda activate ds
pip install -r requirements.txt

This will create a new environment called "ds" with Python version 3.7, activate this environment and install all the packages required (such as Scikit-Learn, Pandas ,etc.). Your commanda line must be pointing at the folder ("cd" to the folder) with the code for the "pip install" command to work.

3. IDE

Development is recommended to be done using PyCharm Community, which can be downloaded here. After opening the code folder with PyCharm, sure to configure the correct conda environment to be the one on which the code runs.

Other information

Updating requirements file

pip freeze > requirements.txt

Project structure

The directory structure of your new project looks like this (please adjust the structure and its description to best fit your project):

├── README.md          <- the top-level README.
│
├── docs               <- resources on Data Science and Data Mining.
│
├── data               <- folder to place data
│
├── notebooks          <- jupyter notebooks
│
├── src                <- code 
│
└── requirements.txt   <- environmnet package requirements

Installation guide in Windows (no administrator rights required)

1. Download Miniconda if you don't have it already (or Anaconda)

2. Make sure to install it for "Just me", and in you user folder

3. Add Anaconda to your PATH environment variable

4. Download the repository from GitHub, if you are familiar with git you can also clone it wherever you want

5. Unpack the code folder, point to it in a Command Line and create the conda environment

6. Install the required libraries using pip

To use PyCharm:

7. Open the project with PyCharm

8. Go to Settings to configure the Python interpreter

9. Choose the existing Conda Environment you created before (named ds)