datatable.mean()

datatable.

mean

(

cols

)

sourcetests

Calculate the mean value for each column from cols.

Parameters

cols

Expr

Input columns.

return

Expr

f-expression having one row, and the same names and number of columns as in cols. The column stypes are float32 for float32 columns, and float64 for all the other numeric types.

except

TypeError

The exception is raised when one of the columns from cols has a non-numeric type.

See Also

• median() – function to calculate median values.

• sd() – function to calculate standard deviation.

Examples

from datatable import dt, f, by df = dt.Frame({'A': [1, 1, 2, 1, 2], 'B': [None, 2, 3,4, 5], 'C': [1, 2, 1, 1, 2]}) df

	A	B	C
	int32	int32	int32
0	1	NA	1
1	1	2	2
2	2	3	1
3	1	4	1
4	2	5	2

5 rows × 3 columns

Get the mean from column A:

df[:, dt.mean(f.A)]

	A
	float64
0	1.4

1 row × 1 column

Get the mean of multiple columns:

df[:, dt.mean([f.A, f.B])]

	A	B
	float64	float64
0	1.4	3.5

1 row × 2 columns

Same as above, but applying to a column slice:

df[:, dt.mean(f[:2])]

	A	B
	float64	float64
0	1.4	3.5

1 row × 2 columns

You can pass in a dictionary with new column names:

df[:, dt.mean({"A_mean": f.A, "C_avg": f.C})]

	A_mean	C_avg
	float64	float64
0	1.4	1.4

1 row × 2 columns

In the presence of by(), it returns the average of each column per group:

df[:, dt.mean({"A_mean": f.A, "B_mean": f.B}), by("C")]

	C	A_mean	B_mean
	int32	float64	float64
0	1	1.33333	3.5
1	2	1.5	3.5

2 rows × 3 columns