Real-world data is rarely ready to use. It has duplicate rows, missing values, inconsistent formats, and columns that need to be split or converted before any analysis makes sense. Data scientists spend a large share of their time on this step, and doing it well is what makes everything after it trustworthy.
You'll start with data wrangling in Pandas: loading files, selecting, filtering, sorting, and grouping data. Then you'll move to cleaning: removing duplicates, dropping columns, standardizing values, and filling in missing data. Finally, you'll watch complete real-world cleaning sessions that combine these techniques on messy datasets.
With clean data in hand, you're ready to explore it visually in Visualization & EDA, the next module in this course.
The wrangling fundamentals: loading CSV and Excel files, selecting rows and columns, filtering with conditions and regex, sorting, modifying a DataFrame, and aggregating with groupby.
A step-by-step cleaning walkthrough: removing duplicates, dropping columns, stripping extra characters, standardizing phone numbers, splitting columns, replacing values, and filling nulls.
Cleaning a messy Olympic athletes dataset end to end: splitting strings into columns, extracting details with regular expressions, and converting columns to datetime and numeric types.
A longer course that ties it together: reading data from CSV, SQL, and Excel, cleaning and transforming it with NumPy and Pandas, and building reports in Jupyter.