Sign in Sign up

Learning path

Exploratory Data Analysis

EDA checklist, missing values and outliers, encoding and scaling, and feature engineering, with animated films and runnable Python. Part of the Data Scientist premium track.

Start with lesson 1
  1. 1

    An EDA Checklist for Any New Dataset

    Exploratory Data Analysis in seven steps: shape, types, missing values, duplicates, distributions, relationships and surprises. A repeatable pandas checklist for every new dataset.

    3 min read
  2. 2

    Missing Values and Outliers: Find, Decide, Fix

    Why data goes missing (MCAR, MAR, MNAR), when to drop or fill, and how to find outliers with the IQR and z-score methods, with pandas code you can…

    3 min read
  3. 3

    Encoding Categories and Scaling Numbers

    Turn text categories into numbers with one-hot, label, ordinal and target encoding, and put numbers on the same scale with standardisation and min-max scaling, using pandas and scikit-learn.

    3 min read
  4. 4

    Feature Engineering: Creating Features Models Can Learn From

    Turn raw columns into powerful features: dates, ratios, aggregates, bins and text lengths. See a weak model jump in accuracy just from better features, with scikit-learn.

    3 min read