Miscellaneous

Confidence Interval

mlpy.percentile_ci_median(x, nboot=1000, alpha=0.025, rseed=0)

Percentile confidence interval for the median of a sample x and unknown distribution.

Input

  • x - [1D numpy array] sample
  • nboot - [integer] (>1) number of resamples
  • alpha - [float] confidence level is 100*(1-2*alpha) (0.0<alpha<1.0)
  • rseed - [integer] random seed

Output

  • ci - (cimin, cimax) confidence interval

Example:

>>> from numpy import *
>>> from mlpy import *
>>> x = array([1,2,4,3,2,2,1,1,2,3,4,3,2])
>>> percentile_ci_median(x, nboot = 100)
(1.8461538461538463, 2.8461538461538463)

Peaks Detection

mlpy.span_pd(x, span)

span peaks detection.

Input

  • x - [1D numpy array float] data
  • span - [odd int] span

Output

  • idx - [1D numpy array integer] peaks indexes

New in version 2.0.7.

Functions from GSL

mlpy.gamma(x)

Gamma Function.

Input

  • x - [float] data

Output

  • gx - [float] gamma(x)
mlpy.fact(x)

Factorial x!. The factorial is related to the gamma function by x! = gamma(x+1)

Input

  • x - [int] data

Output

  • fx - [float] factorial x!
mlpy.quantile(x, f)

Quantile value of sorted data. The elements of the array must be in ascending numerical order. The quantile is determined by the f, a fraction between 0 and 1. The quantile is found by interpolation, using the formula: quantile = (1 - delta) x_i + delta x_{i+1} where i is floor((n - 1)f) and delta is (n-1)f - i.

Input

  • x - [1D numpy array float] sorted data
  • f - [float] fraction between 0 and 1

Output

  • q - [float] quantile
mlpy.cdf_gaussian_P(x, sigma)

Cumulative Distribution Functions (CDF) P(x) for the Gaussian distribution.

Input

  • x - [float] data
  • sigma - [float] standard deviation

Output

  • p - [float]

New in version 2.0.2.

Other

mlpy.away(a, b, d)

Given numpy 1D array a and numpy 1D array b compute c = { bi : | bi - aj | > d for each i, j}

Input

  • a - [1D numpy array float]
  • b - [1D numpy array float]
  • d - [double]

Output

  • c - [1D numpy array float]

New in version 2.0.3.

mlpy.is_power(n, b)

Return True if ‘n’ is power of ‘b’, False otherwise.

New in version 2.0.6.

mlpy.next_power(n, b)

Returns the smallest integer, greater than or equal to ‘n’ which can be obtained as power of ‘b’.

New in version 2.0.6.