base-funs

In [1]:

%run nb_helpers.py

from datar.all import *

debug_kwargs = {'prefix': '\n', 'sep': f'\n{"-" * 20}\n'}
nb_header(
    cut, diff, identity, expand_grid, outer, 
    make_names, make_unique, rank,
)

%run nb_helpers.py from datar.all import * debug_kwargs = {'prefix': '\n', 'sep': f'\n{"-" * 20}\n'} nb_header( cut, diff, identity, expand_grid, outer, make_names, make_unique, rank, )

Try this notebook on binder.

★ cut

¶

Cut a numeric vector into bins¶

Args:¶

  x: A numeric vector
  breaks: The breaks
  labels: The labels
  include_lowest: Whether to include the lowest value
  right: Whether to include the rightmost value
  dig_lab: The number of digits for labels
  ordered_result: Whether to return an ordered factor

Returns:¶

  The factor vector

★ diff

¶

Difference of a numeric vector¶

Args:¶

  x: A numeric vector
  lag: The lag to use. Could be negative.
    It always calculates x[lag:] - x[:-lag] even when lag is
    negative

  differences: The order of the difference

Returns:¶

  An array of x[lag:] – x[:-lag].
  If differences > 1, the rule applies differences times on x

★ identity

¶

Identity function¶

Args:¶

  x: A numeric vector

Returns:¶

  The same vector

★ expand_grid

¶

Expand a grid¶

Args:¶

  x: A numeric vector
  *args: Additional numeric vectors
  **kwargs: Additional keyword arguments

Returns:¶

  The expanded grid

★ outer

¶

Outer product of two vectors¶

Args:¶

  x: A numeric vector
  y: A numeric vector
  fun: The function to handle how the result of the elements from
    the first and second vectors should be computed.
    The function has to be vectorized at the second argument, and
    return the same shape as y.

Returns:¶

  The outer product

★ make_names

¶

Make names for a vector¶

Args:¶

  names: character vector to be coerced to syntactically valid names.
    This is coerced to character if necessary.

  unique: Whether to make the names unique

Returns:¶

  The names

★ make_unique

¶

Make a vector unique¶

Args:¶

  names: a character vector

Returns:¶

  The unique vector

★ rank

¶

Rank a numeric vector¶

Args:¶

  x: A numeric vector
  na_last: Whether to put NA at the end
  ties_method: The method to handle ties. One of "average", "first",
    "last", "random", "max", "min"

Returns:¶

  The ranks

In [2]:

debug(
    cut(seq(1,10), 3), 
    diff([1, 2, 3]),
    identity(1.23),
    expand_grid([1,2], [3,4]),
    outer([1,2], [3,4]),
    make_names([1, 2, 3]),
    make_unique([1, 1, 1]),
    rank([3, 4, 1, -1]),
    **debug_kwargs
)

debug( cut(seq(1,10), 3), diff([1, 2, 3]), identity(1.23), expand_grid([1,2], [3,4]), outer([1,2], [3,4]), make_names([1, 2, 3]), make_unique([1, 1, 1]), rank([3, 4, 1, -1]), **debug_kwargs )

[2022-12-02 13:26:45][datar][WARNING] New names:
[2022-12-02 13:26:45][datar][WARNING] * '_1' -> '__0'
[2022-12-02 13:26:45][datar][WARNING] * '_2' -> '__1'
[2022-12-02 13:26:45][datar][WARNING] * '_3' -> '__2'
[2022-12-02 13:26:45][datar][WARNING] New names:
[2022-12-02 13:26:45][datar][WARNING] * '_1' -> '__0'
[2022-12-02 13:26:45][datar][WARNING] * '_1' -> '__1'
[2022-12-02 13:26:45][datar][WARNING] * '_1' -> '__2'

cut(seq(1,10), 3)
--------------------
[(0.99, 4.0], (0.99, 4.0], (0.99, 4.0], (0.99, 4.0], (4.0, 7.0], (4.0, 7.0], (4.0, 7.0], (7.0, 10.0], (7.0, 10.0], (7.0, 10.0]]
Categories (3, interval[float64, right]): [(0.99, 4.0] < (4.0, 7.0] < (7.0, 10.0]]

diff([1, 2, 3])
--------------------
array([1, 1])

identity(1.23)
--------------------
1.23

expand_grid([1,2], [3,4])
--------------------
   _VAR_0  _VAR_1
  <int64> <int64>
0       1       3
1       1       4
2       2       3
3       2       4

outer([1,2], [3,4])
--------------------
        0       1
  <int64> <int64>
0       3       4
1       6       8

make_names([1, 2, 3])
--------------------
['__0', '__1', '__2']

make_unique([1, 1, 1])
--------------------
['__0', '__1', '__2']

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