datamash

datamash [options] operation column [operation column...]

datamash performs basic numeric, textual, and statistical operations on input data from the command line. It's designed for quick data analysis tasks that would otherwise require scripting or statistical software, supporting operations like sum, mean, median, standard deviation, variance, and more.Input is read from stdin or files, with columns separated by whitespace or a specified delimiter. The tool can group data by fields and compute aggregate statistics for each group, similar to SQL's GROUP BY functionality.datamash is part of the GNU project and excels at one-liners for data exploration. It's commonly used in pipelines to analyze CSV files, log data, or any tabular text data. The tool can handle both numeric and textual operations, including counting unique values, string operations, and random sampling.

-R, --round digits

Round numeric output to specified decimals

Ignore NA and NaN values

Use specified field separator

Group by specified fields

First line is header (both input and output)

Sort the input before grouping (required if input is not already sorted by the group fields)

Use whitespace (one or more spaces/tabs) as field separator

Print entire input line before the operation results

sum, min, max, mean, median

Basic statistics

Population/sample standard deviation

Counting and grouping

Selection operations

The decimal separator follows the current locale (period in the C locale, comma in e.g. German locales). Grouping (-g) expects sorted input unless -s is given. Column numbering starts at 1.

awk(1), cut(1), sort(1)

Homepage · Documentation