Python and R for Data Science (O'Reilly Live Online Training)

Setup

Python

If you don't have Python pre-installed on your machine, you can download it from python.org. After it's installed you need to set up a virtual environment:

python3 -m venv venv
source venv/bin activate

And then you can either install all the packages required with pip individually, or use the following command:

pip install -r requirements.md

To start using JupyterLab:

jupyter lab

If you want to use a text editor, a recommended tool is VSCode. While most of the R work will be in RStudio, we'll use JupyterLab for a small section, so you need to install the IRKernel. For that follow the official setup instructions.

R

You can download R from CRAN. After this you should install RStudio from the official website. After this create a new R Project (select use existing directory, use the one where you download this repository) and install the packages individually with install.packages().

Required packages

Python

• jupyterlab (Python and R Kernels)
• pandas
• scikit-learn
• yellowbrick
• missingno
• seaborn
• opencv-python
• scikit-image
• nltk (download data as well)
• spaCy (download model with python -m spacy download en_core_web_sm)
• flask (optional)
• mlflow (optional)

R

• ggplot2
• dplyr
• prophet
• xts
• leaflet
• shiny
• flexdashboard (optional)
• shinydashboard (optional)
• sdmbench (optional)

Datasets

• diamonds.csv (built-in from ggplot2)
• boston housing (built-in from scikit-learn)
• wine (built in from scikit-learn)
• fires.csv (from Kaggle)
• reviews.csv (from UCSD)
• image_000001.jpg (flowers102 from University of Oxford)
• starwars.csv (built-in from dplyr)
• temps.csv (from Machine Learning Mastery Github)
• quakes (built-in from RStudio)
• data_for_ml.csv (intermediate dataset for case study - processed fires.csv)