pyinterp.TDigestFloat64

class pyinterp.TDigestFloat64(self, values: ndarray[dtype=float64, device='cpu'], weights: ndarray[dtype=float64, device='cpu'] | None = None, axis: collections.abc.Sequence[int] | None = None, compression: int | None = 100)

Bases: object

T-Digest for incremental quantile estimation.

Computes quantiles using the t-digest algorithm which provides accurate estimates, especially at the tails of the distribution. Supports parallel and online computation with arbitrary weights.

Reference: Computing Extremely Accurate Quantiles Using t-Digests tdunning/t-digest

Parameters:

• values – Input array of values.

• weights – Optional array of weights (same shape as values).

• axis – Optional axis or axes along which to compute quantiles.

• compression – Compression parameter controlling accuracy vs memory tradeoff. Higher values provide better accuracy but use more memory. Typical values: 100-1000. Default is 100.

Example

>>> import numpy as np
>>> import pyinterp
>>> data = np.random.randn(10000)
>>> tdigest = pyinterp.TDigest(data)
>>> median = tdigest.quantile(0.5)
>>> print(f"Median: {median}")

Compute multiple quantiles

>>> quantiles = tdigest.quantile(np.array([0.25, 0.5, 0.75]))
>>> print(f"Q25, Q50, Q75: {quantiles}")

Compute along axis

>>> data_2d = np.random.randn(100, 50)
>>> tdigest_axis = pyinterp.TDigest(data_2d, axis=[0])
>>> medians = tdigest_axis.quantile(0.5)
>>> print(f"Medians shape: {medians.shape}")

Public Methods

count(self)	Return the count of samples.
max(self)	Return the maximum of samples.
mean(self)	Return the mean of samples.
min(self)	Return the minimum of samples.
quantile(-> numpy.ndarray[dtype=float64])	Overloaded function.
sum_of_weights(self)	Return the sum of weights.

Special Methods

__copy__(self)	Create a copy of this object.
__getstate__(self)	Get the state for pickling.
__iadd__(self, other)	Override the default behavior of the += operator.
__new__(*args, **kwargs)
__setstate__(self, state)	Set the state for unpickling.