kwarray.distributions

Defines data structures for efficient repeated sampling of specific distributions (e.g. Normal, Uniform, Binomial) with specific parameters.

Inspired by ~/code/imgaug/imgaug/parameters.py

Similar Libraries:

Todo

  • [ ] change sample shape to just a single num.

  • [ ] Some Distributions will output vectors. Maybe we could just postpend the dimensions?

  • [ ] Expose as kwstats?

Module Contents

Classes

Uniform

Defaults to a uniform distribution over floats between 0 and 1

Exponential

Example

Constant

Example

DiscreteUniform

Uniform distribution over integers.

Normal

 

>>> self = Normal(mean=100, rng=0)

Bernoulli

self = Normal()

Binomial

self = Normal()

Categorical

Example

TruncNormal

A truncated normal distribution.
class kwarray.distributions.Uniform(low=0, high=1, rng=None)

Bases: Distribution

Defaults to a uniform distribution over floats between 0 and 1

Example

>>> self = Uniform(rng=0)
>>> self.sample()
0.548813...
>>> float(self.sample(1))
0.7151...
Benchmark:
>>> import ubelt as ub
>>> self = Uniform()
>>> for timer in ub.Timerit(100, bestof=10):
>>>     with timer:
>>>         [self() for _ in range(100)]
>>> for timer in ub.Timerit(100, bestof=10):
>>>     with timer:
>>>         self(100)
sample(self, *shape)
classmethod coerce(cls, arg)
class kwarray.distributions.Exponential(scale=1, rng=None)

Bases: Distribution

Example

>>> self = Exponential(rng=0)
>>> # xdoctest: +REQUIRES(--show)
>>> import kwplot
>>> kwplot.autompl()
>>> kwplot.figure(fnum=1, doclf=True)
>>> self._show(500, bins=25)
sample(self, *shape)
class kwarray.distributions.Constant(value=0, rng=None)

Bases: Distribution

Example

>>> self = Constant(42, rng=0)
>>> self.sample()
42
>>> self.sample(3)
array([42, 42, 42])
sample(self, *shape)
class kwarray.distributions.DiscreteUniform(min=0, max=1, rng=None)

Bases: Distribution

Uniform distribution over integers.

Parameters

  • min (int) – inclusive minimum

  • max (int) – exclusive maximum

Example

>>> self = DiscreteUniform.coerce(4)
>>> self.sample(100)
sample(self, *shape)
classmethod coerce(cls, arg, rng=None)
class kwarray.distributions.Normal(mean=0, std=1, rng=None)

Bases: Distribution

>>> self = Normal(mean=100, rng=0)
>>> self.sample()
>>> self.sample(100)
sample(self, *shape)
class kwarray.distributions.Bernoulli(p=0.5, rng=None)

Bases: Distribution

self = Normal() self.sample() self.sample(1)

sample(self, *shape)
classmethod coerce(cls, arg)
class kwarray.distributions.Binomial(n=1, p=0.5, rng=None)

Bases: Distribution

self = Normal() self.sample() self.sample(1)

sample(self, *shape)
class kwarray.distributions.Categorical(categories, weights=None, rng=None)

Bases: Distribution

Example

>>> categories = [3, 5, 1]
>>> weights = [.05, .5, .45]
>>> self = Categorical(categories, weights, rng=0)
>>> self.sample()
5
>>> list(self.sample(2))
[1, 1]
>>> self.sample(2, 3)
array([[5, 5, 1],
       [5, 1, 1]])
sample(self, *shape)
class kwarray.distributions.TruncNormal(mean=0, std=1, low=- np.inf, high=np.inf, rng=None)

Bases: Distribution

A truncated normal distribution.

A normal distribution, but bounded by low and high values. Note this is much different from just using a clipped normal.

Parameters

  • mean (float) – mean of the distribution

  • std (float) – standard deviation of the distribution

  • low (float) – lower bound

  • high (float) – upper bound

  • rng (np.random.RandomState)

Example

>>> self = TruncNormal(rng=0)
>>> self()  # output of this changes before/after scipy version 1.5
...0.1226...

Example

>>> low = -np.pi / 16
>>> high = np.pi / 16
>>> std = np.pi / 8
>>> self = TruncNormal(low=low, high=high, std=std, rng=0)
>>> shape = (3, 3)
>>> data = self(*shape)
>>> print(ub.repr2(data, precision=5))
np.array([[ 0.01841,  0.0817 ,  0.0388 ],
          [ 0.01692, -0.0288 ,  0.05517],
          [-0.02354,  0.15134,  0.18098]], dtype=np.float64)
_update_internals(self)
sample(self, *shape)