openest.models.spline_model module

class openest.models.spline_model.SplineModel(xx_is_categorical=False, xx=None, conditionals=None, scaled=True)

Bases: openest.models.univariate_model.UnivariateModel, openest.models.memoizable.MemoizableUnivariate

Model Spline File

Each line in a model spline file represents a polynomial segment in log-probability space. The format is as follows:

spp1
<x>,<y0>,<y1>,<a0>,<a1>,<a2>
...

Each line describes a segment of a probability distribution of y, conditional on x = <x>. The segment spans from <y0> to <y1>, where the lowest value of <y0> may be -inf, and the highest value of <y1> may be inf. The <x> values may also be categorical or numerical. If they are numerical, it is assumed that these values represent samples of a smoothly varying function (a cubic spline in every y).

The values <a0>, <a1> and <a2> are the polynomial coefficients in y (with quadratic coefficients, only normal or exponential tails are possible). The final segment of the probability function is:

exp(a0 + a1 y + a2 y2)

Parameters:

• xx_is_categorical (bool) –
• xx (list-like) –
• conditionals –
• scaled (bool) –

add_conditional(x, conditional)

cdf(xx, yy)

static combine(one, two)

copy()

static create_gaussian(xxs, order=None, xx_is_categorical=True)

xxs should be a dictionary of the form {x: (mean, variance)}.

static create_single(xxs, y0s, y1s, coeffss, order=None, xx_is_categorical=True)

draw_sample(x=None)

Produce a sample value of y from the conditional distribution.

eval_pval(x, p, threshold=0.001)

Inverse CDF Evaluation

Returns the value of $y$ that corresponds to a given p-value: $F^{-1}(p | x)$.

eval_pval_index(ii, p, threshold=0.001)

filter_x(xx)

static from_ddp(ddp_model, limits)

get_conditional(x)

get_mean(x=None)

E[Y | X]

get_sdev(x=None)

sqrt Var[Y | X]

get_xx()

Listing conditional values

Provide a list of all sampled conditional values.

init_from_spline_file(file, delimiter, status_callback=None)

interpolate_x(newxx)

Determines whether argument newxx a subset of index xx.

is_gaussian(x=None)

kind()

static merge(models)

neginf = -inf

posinf = inf

recategorize_x(oldxx, newxx)

Construct a new model with categorical x values ‘newxx’, using the conditionals currently assigned to categorical x values ‘oldxx’.

samples = 1000

scale_p(a)

Raise the distribution to the power ‘a’ and rescales.

Returns:modifies this model and returns it Return type:self

scale_y(a)

Rescaling of the Parameter Dimension

Produces a new conditional PDF with the $y$ dimension scaled by a constant: $p(z | x) = p( rac{y}{a} | x)$.

to_ddp(ys=None)

to_points_at(x, ys)

Conditional Probability Density Evaluation

Returns unscaled probability density values for given values of $x$ and $y$: $f(y | x)$.

write(file, delimiter)

write_file(filename, delimiter)

write_gaussian(file, delimiter)

write_gaussian_plus(file, delimiter)

class openest.models.spline_model.SplineModelConditional(y0s=None, y1s=None, coeffs=None)

add_segment(y0, y1, coeffs)

approximate_mean(limits)

static approximate_sum(conditionals)

static ascinv(y, func, minx, maxx, threshold)

cdf(yy)

convolve(other)

copy()

draw_sample()

evaluate(ii, y)

find_mode()

static find_nearest(array, value, within)

gaussian_mean(ii)

gaussian_sdev(ii)

get_pval(p, threshold=0.001)

is_gaussian()

static make_conditional_from_spline(spline, limits)

static make_gaussian(y0, y1, mean, var)

static make_single(y0, y1, coeffs)

nongaussian_x2px(ii)

nongaussian_xpx(ii)

partial_cdf(ii, y1)

static propose_grid(conditionals)

rescale()

rough_limits()

rough_span()

scale(factor)

scale_p(a)

scale_y(a)

segment_max(jj)

size()

to_points(ys)