Pandas Cleaning

Real data often has missing values, duplicate lines, and the wrong data type so it needs to be cleaned up before it's analyzed. Pandas provides dropna() to throw out blank value rows, fillna() to fill in (e.g. with average columns), and drop_duplicates() to delete duplicates. The to_numeric() function is useful when the number column is read as text. A neat data cleanup becomes the foundation for reliable analysis and model results.