TimeRolling¶
Wrapper for measuring probability distributions over a period of time.
Parameters¶
-
dist (river.proba.base.Distribution)
A distribution.
-
period (datetime.timedelta)
A period of time.
Attributes¶
-
n_samples
The number of observed samples.
Examples¶
>>> import datetime as dt
>>> from river import proba
>>> X = ['red', 'green', 'green', 'blue']
>>> days = [1, 2, 3, 4]
>>> dist = proba.TimeRolling(
... proba.Multinomial(),
... period=dt.timedelta(days=2)
... )
>>> for x, day in zip(X, days):
... dist = dist.update(x, dt.datetime(2019, 1, day))
... print(dist)
... print()
P(red) = 1.000
<BLANKLINE>
P(red) = 0.500
P(green) = 0.500
<BLANKLINE>
P(green) = 1.000
P(red) = 0.000
<BLANKLINE>
P(green) = 0.500
P(blue) = 0.500
P(red) = 0.000
<BLANKLINE>
Methods¶
revert
Reverts the parameters of the distribution for a given observation.
Parameters
- x
- w – defaults to
1
update
Updates the parameters of the distribution given a new observation.
Parameters
- x
- t (datetime.datetime)