This is a GitHub project that easily allows you to create a report from a pandas DataFrame. The package will present the following measures in an interactive HTML report, which is used to evaluate the data at hand for a data science project:
- Essentials, such as type, unique values, and missing values
- Quantile statistics, such as minimum value, Q1, median, Q3, maximum, range, and interquartile range
- Descriptive statistics such as mean, mode, standard deviation, sum, median absolute deviation, the coefficient of variation, kurtosis, and skewness
- Most frequent values
- Histograms
- Correlations highlighting highly correlated variables, and Spearman and Pearson matrixes
Here is all the information about this package:
- Website: https://github.com/pandas-profiling/pandas-profiling
- Version at the time of print: 1.4.1
- Suggested install command: pip install pandas-profiling