📓
Dataprocessing
  • Preface
  • Essential Libraries
  • Introduction
  • Tidy data
  • Vectorized solutions
  • Load data
    • CSV
    • XML
    • SQL
    • json
  • Data Inspection
    • Dealing with missing values
    • Cleaning data
    • Normalization
  • Data Preparation
    • Tidy data
    • Combine data
    • Reshaping
      • Reshape with melt
  • Data exploration
  • Data visualisation
    • Panel basics
  • Plotten van kaarten
  • Data analysis
    • Statistics in Python
  • Databases
    • Database concepts
  • studycases
    • Montly sales data
      • Make it interactive using panel
    • Combine data to investigate a hypothesis
    • Twitter analysis using sql
  • Missing time related data
  • Plotting F1 race data
Powered by GitBook
On this page

Was this helpful?

Vectorized solutions

Why we should always try the vectorised implementation

PreviousTidy dataNextLoad data

Last updated 2 years ago

Was this helpful?

Data Science we often deal with big data. Data manipulations can easily take up loads of computer time and memory when not efficiently coded. In order to fully take advantage of computation power of computers, the state of art of implementation of algorithm is vectorising all the manipulations. This allows you to achieve parallelised computation using multiple cpu's at once. The libraries pandas and numpy allow vectorised manipulations. This makes the computations easily 2500 times faster. So stop looping over rows and columns. Use the pandas methods!