/data_science_exercise

Technical exercise

Primary LanguageJupyter Notebook

Data science technical exercise

Instructions

  • The report containing the analysis can be found at reports/report.md
  • In order to reproduce the results, the setup.sh script can be user:
    • Add Data scientist exercise.db to the data folder
    • ./setup.sh build_venv : Creates a local Python environment from the requirements.txt file
    • ./setup.sh clean_data : Cleans the data, prints general metrics for the dataset, and generates plots that are saved in reports/figs/raw_logs_eda and reports/figs/clean_logs_eda
    • ./setup.sh experiment_report : Prints the results of the statistical experiments and generates plots for the bootstrapped distributions (reports/figs/test)
    • ./setup.sh estimate_percentage : Prints the results of the percentage increase analysus and generates plots for the distributions (reports/figs/percentage)

Content

  • data : Folder for the SQLite database
  • docs : Instructions
  • notebooks : Scratchpad notebooks for simple experiments
  • reports : Final report
  • src : Python scripts