base-arithmetic

In [1]:

%run nb_helpers.py

from datar.all import *

debug_kwargs = {'prefix': '\n', 'sep': f'\n{"-" * 20}\n'}
nb_header(
    sum, prod, mean, median, min, max, var, pmin, pmax,
    round, sqrt, abs, sign, trunc, ceiling, floor, signif,
    log, exp, log2, log10, log1p, cov, scale, col_sums,
    row_sums, col_means, row_means, col_sds, row_sds,
    col_medians, row_medians, quantile, sd, weighted_mean
)

%run nb_helpers.py from datar.all import * debug_kwargs = {'prefix': '\n', 'sep': f'\n{"-" * 20}\n'} nb_header( sum, prod, mean, median, min, max, var, pmin, pmax, round, sqrt, abs, sign, trunc, ceiling, floor, signif, log, exp, log2, log10, log1p, cov, scale, col_sums, row_sums, col_means, row_means, col_sds, row_sds, col_medians, row_medians, quantile, sd, weighted_mean )

Try this notebook on binder.

★ sum_

¶

Compute the sum of a vector¶

Args:¶

  x: A numeric vector
  na_rm: Whether to remove NA values

Returns:¶

  The sum of the vector

★ prod

¶

Compute the product of a vector¶

Args:¶

  x: A numeric vector
  na_rm: Whether to remove NA values

Returns:¶

  The product of the vector

★ mean

¶

Compute the mean of a vector¶

Args:¶

  x: A numeric vector
  na_rm: Whether to remove NA values

Returns:¶

  The mean of the vector

★ median

¶

Compute the median of a vector¶

Args:¶

  x: A numeric vector
  na_rm: Whether to remove NA values

Returns:¶

  The median of the vector

★ min_

¶

Compute the minimum of a vector¶

Args:¶

  x: A numeric vector
  na_rm: Whether to remove NA values

Returns:¶

  The minimum of the vector

★ max_

¶

Compute the maximum of a vector¶

Args:¶

  x: A numeric vector
  na_rm: Whether to remove NA values

Returns:¶

  The maximum of the vector

★ var

¶

Compute the variance of a vector¶

Args:¶

  x: A numeric vector
  y: None or a vector, matrix or data frame with
    compatible dimensions to x. The default is equivalent to
    y = x

  na_rm: Whether to remove NA values
  ddof: The degrees of freedom

Returns:¶

  The variance of the vector

★ pmin

¶

Returns the (regular or Parallel) maxima and minima of the input¶

  values.

Args:¶

  x: A numeric vector
  more: One or more values
  na_rm: Whether to remove NA values

Returns:¶

  The minimum of the vector and the values

★ pmax

¶

Returns the (regular or Parallel) maxima and minima of the input¶

  values.

Args:¶

  x: A numeric vector
  more: One or more values
  na_rm: Whether to remove NA values

Returns:¶

  The maximum of the vector and the values

★ round_

¶

Round the values of a vector¶

Args:¶

  x: A numeric vector
  digits: The number of digits to round to

Returns:¶

  The rounded values

★ sqrt

¶

Compute the square root of a vector¶

Args:¶

  x: A numeric vector

Returns:¶

  The square root of the vector

★ abs_

¶

Compute the absolute value of a vector¶

Args:¶

  x: A numeric vector

Returns:¶

  The absolute values of the vector

★ sign

¶

Compute the sign of a vector¶

Args:¶

  x: A numeric vector

Returns:¶

  The signs of the vector

★ trunc

¶

Truncate the values of a vector¶

Args:¶

  x: A numeric vector

Returns:¶

  The truncated values

★ ceiling

¶

Round up to the nearest integer¶

Args:¶

  x: The value to be rounded up

Returns:¶

  The rounded up value

★ floor

¶

Round down to the nearest integer¶

Args:¶

  x: The value to be rounded down

Returns:¶

  The rounded down value

★ signif

¶

Round the values of a vector to a given number of significant digits¶

Args:¶

  x: A numeric vector
  digits: The number of significant digits to round to

Returns:¶

  The rounded values

★ log

¶

Compute the logarithm of a vector¶

Args:¶

  x: A numeric vector
  base: The base of the logarithm

Returns:¶

  The logarithm values

★ exp

¶

Compute the exponential of a vector¶

Args:¶

  x: A numeric vector

Returns:¶

  The exponential values

★ log2

¶

Compute the base-2 logarithm of a vector¶

Args:¶

  x: A numeric vector

Returns:¶

  The logarithm values

★ log10

¶

Compute the base 10 logarithm of a vector¶

Args:¶

  x: A numeric vector

Returns:¶

  The logarithm values

★ log1p

¶

Compute the logarithm of one plus a vector¶

Args:¶

  x: A numeric vector

Returns:¶

  The logarithm values

★ cov

¶

Compute pairwise covariance between two variables¶

Args:¶

  x: a numeric vector, matrix or data frame.
  y: None or a vector, matrix or data frame with
    compatible dimensions to x. The default is equivalent to
    y = x

  na_rm: If True, remove missing values before computing
    the covariance.

  ddof: The denominator degrees of freedom.

Returns:¶

  The covariance matrix

★ scale

¶

Center and/or scale the data¶

Args:¶

  x: A numeric vector
  center: Whether to center the data
  scale_: Whether to scale the data

Returns:¶

  The scaled data

★ col_sums

¶

Compute the column sums of a matrix¶

Args:¶

  x: A numeric matrix
  na_rm: Whether to remove NA values

Returns:¶

  The column sums of the matrix

★ row_sums

¶

Compute the row sums of a matrix¶

Args:¶

  x: A numeric matrix
  na_rm: Whether to remove NA values

Returns:¶

  The row sums of the matrix

★ col_means

¶

Compute the column means of a matrix¶

Args:¶

  x: A numeric matrix
  na_rm: Whether to remove NA values

Returns:¶

  The column means of the matrix

★ row_means

¶

Compute the row means of a matrix¶

Args:¶

  x: A numeric matrix
  na_rm: Whether to remove NA values

Returns:¶

  The row means of the matrix

★ col_sds

¶

Compute the column standard deviations of a matrix¶

Args:¶

  x: A numeric matrix
  na_rm: Whether to remove NA values

Returns:¶

  The column standard deviations of the matrix

★ row_sds

¶

Compute the row standard deviations of a matrix¶

Args:¶

  x: A numeric matrix
  na_rm: Whether to remove NA values

Returns:¶

  The row standard deviations of the matrix

★ col_medians

¶

Compute the column medians of a matrix¶

Args:¶

  x: A numeric matrix
  na_rm: Whether to remove NA values

Returns:¶

  The column medians of the matrix

★ row_medians

¶

Compute the row medians of a matrix¶

Args:¶

  x: A numeric matrix
  na_rm: Whether to remove NA values

Returns:¶

  The row medians of the matrix

★ quantile

¶

Compute the quantiles of a vector¶

Args:¶

  x: A numeric vector
  probs: The probabilities to use

Returns:¶

  The quantiles of the vector

★ sd

¶

Compute the standard deviation of a vector¶

Args:¶

  x: A numeric vector
  na_rm: Whether to remove NA values

Returns:¶

  The standard deviation of the vector

★ weighted_mean

¶

Compute the weighted mean of a vector¶

Args:¶

  x: A numeric vector
  w: The weights to use
  na_rm: Whether to remove NA values

Returns:¶

  The weighted mean of the vector

In [2]:

debug(
    sum([1,2,4,6]),
    prod([1,2,4,6]),
    mean([1,2,4,6]),
    median([1,2,4,6]),
    min([1,2,4,6]),
    max([1,2,4,6]),
    var([1,2,4,6]),
    pmin([1,4], [2,3]),
    pmax([1,4], [2,3]),
    **debug_kwargs
)

debug( sum([1,2,4,6]), prod([1,2,4,6]), mean([1,2,4,6]), median([1,2,4,6]), min([1,2,4,6]), max([1,2,4,6]), var([1,2,4,6]), pmin([1,4], [2,3]), pmax([1,4], [2,3]), **debug_kwargs )

sum([1,2,4,6])
--------------------
13

prod([1,2,4,6])
--------------------
48

mean([1,2,4,6])
--------------------
3.25

median([1,2,4,6])
--------------------
3.0

min([1,2,4,6])
--------------------
1

max([1,2,4,6])
--------------------
6

var([1,2,4,6])
--------------------
4.916666666666667

pmin([1,4], [2,3])
--------------------
array([1, 3])

pmax([1,4], [2,3])
--------------------
array([2, 4])

In [3]:

debug(
    round([1.4, 1.5]),
    sqrt([1.1, 2.1]),
    abs([1, -1]),
    sign([10, -10]),
    trunc([1.1, 2.1]),
    ceiling([1.1, 2.1]),
    floor([1.1, 2.1]),
    signif(3.14567e-10, 3),
    **debug_kwargs
)

debug( round([1.4, 1.5]), sqrt([1.1, 2.1]), abs([1, -1]), sign([10, -10]), trunc([1.1, 2.1]), ceiling([1.1, 2.1]), floor([1.1, 2.1]), signif(3.14567e-10, 3), **debug_kwargs )

round([1.4, 1.5])
--------------------
array([1., 2.])

sqrt([1.1, 2.1])
--------------------
array([1.04880885, 1.44913767])

abs([1, -1])
--------------------
array([1, 1])

sign([10, -10])
--------------------
array([ 1, -1])

trunc([1.1, 2.1])
--------------------
array([1., 2.])

ceiling([1.1, 2.1])
--------------------
array([2., 3.])

floor([1.1, 2.1])
--------------------
array([1., 2.])

signif(3.14567e-10, 3)
--------------------
array(3.15e-10)

In [4]:

debug(
    log(exp(2)),
    exp(2),
    log2(4),
    log10(100),
    log1p(exp(1)-1),
    cov([1, 2, 3], [3, 2, 1]),
    scale([1, 2, 3]),
    **debug_kwargs
)

debug( log(exp(2)), exp(2), log2(4), log10(100), log1p(exp(1)-1), cov([1, 2, 3], [3, 2, 1]), scale([1, 2, 3]), **debug_kwargs )

log(exp(2))
--------------------
2.0

exp(2)
--------------------
7.38905609893065

log2(4)
--------------------
2.0

log10(100)
--------------------
2.0

log1p(exp(1)-1)
--------------------
1.0

cov([1, 2, 3], [3, 2, 1])
--------------------
-1.0

scale([1, 2, 3])
--------------------
array([-1.,  0.,  1.])

In [5]:

# column and row stats

df = tribble(f.v1, f.v2, f.v3, *rnorm(15))
debug(
    df,
    col_sums(df),
    row_sums(df),
    col_means(df),
    row_means(df),
    col_sds(df),
    row_sds(df),
    col_medians(df),
    row_medians(df),
    **debug_kwargs,
)

# column and row stats df = tribble(f.v1, f.v2, f.v3, *rnorm(15)) debug( df, col_sums(df), row_sums(df), col_means(df), row_means(df), col_sds(df), row_sds(df), col_medians(df), row_medians(df), **debug_kwargs, )

df
--------------------
         v1        v2        v3
  <float64> <float64> <float64>
0  0.067273  0.015561 -0.633373
1 -0.911285  1.526871  0.540873
2  1.140124 -0.596385  1.933209
3 -0.147867 -0.945199  0.994281
4 -2.492679 -1.536482 -0.473660

col_sums(df)
--------------------
v1   -2.344435
v2   -1.535634
v3    2.361329
dtype: float64

row_sums(df)
--------------------
0   -0.550539
1    1.156458
2    2.476947
3   -0.098786
4   -4.502821
dtype: float64

col_means(df)
--------------------
v1   -0.468887
v2   -0.307127
v3    0.472266
dtype: float64

row_means(df)
--------------------
0   -0.183513
1    0.385486
2    0.825649
3   -0.032929
4   -1.500940
dtype: float64

col_sds(df)
--------------------
v1    1.348229
v2    1.169380
v3    1.064046
dtype: float64

row_sds(df)
--------------------
0    0.390447
1    1.226483
2    1.293786
3    0.974835
4    1.009978
dtype: float64

col_medians(df)
--------------------
v1   -0.147867
v2   -0.596385
v3    0.540873
dtype: float64

row_medians(df)
--------------------
0    0.015561
1    0.540873
2    1.140124
3   -0.147867
4   -1.536482
dtype: float64

In [6]:

debug(
    quantile([1, 2, 3, 4, 5]),
    quantile([1, 2, 3, 4, 5], [0, 1]),
    sd([1, 2, 3, 4, 5]),
    weighted_mean([1, 2, 3, 4, 5]),
    weighted_mean([1, 2, 3, 4, 5], [5, 4, 3, 2, 1]),
    **debug_kwargs,
)

debug( quantile([1, 2, 3, 4, 5]), quantile([1, 2, 3, 4, 5], [0, 1]), sd([1, 2, 3, 4, 5]), weighted_mean([1, 2, 3, 4, 5]), weighted_mean([1, 2, 3, 4, 5], [5, 4, 3, 2, 1]), **debug_kwargs, )

quantile([1, 2, 3, 4, 5])
--------------------
array([1., 2., 3., 4., 5.])

quantile([1, 2, 3, 4, 5], [0, 1])
--------------------
array([1, 5])

sd([1, 2, 3, 4, 5])
--------------------
1.5811388300841898

weighted_mean([1, 2, 3, 4, 5])
--------------------
3.0

weighted_mean([1, 2, 3, 4, 5], [5, 4, 3, 2, 1])
--------------------
2.3333333333333335