16343364439

Committed 17 Jul 2025 11:02AM UTC coverage: 77.939% (+0.8%) from 77.136%

Build # 16343364439

Build Type

push

github

Committed by

avivajpeyi

Commit Message

fix: update inference result saving/loading

Run Details

76 of 103 branches covered (73.79%)

Branch coverage included in aggregate %.

125 of 130 new or added lines in 12 files covered. (96.15%)

14 existing lines in 4 files now uncovered.

892 of 1139 relevant lines covered (78.31%)

1.57 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

78.46

/src/log_psplines/plotting/utils.py

import dataclasses

import arviz as az
import jax.numpy as jnp
import numpy as np

from ..datatypes import Periodogram
from ..psplines import LogPSplines

__all__ = ["unpack_data"]


@dataclasses.dataclass
class PlottingData:
    freqs: np.ndarray = None
    pdgrm: np.ndarray = None
    model: np.ndarray = None
    ci: np.ndarray = None

    @property
    def n(self):
        if self.freqs is not None:
            return len(self.freqs)
        elif self.pdgrm is not None:
            return len(self.pdgrm)
        elif self.model is not None:
            return len(self.model)
        else:
            raise ValueError("No data to get length from.")


def unpack_data(
    pdgrm: Periodogram = None,
    spline_model: LogPSplines = None,
    weights=None,
    yscalar=1.0,
    use_uniform_ci=True,
    use_parametric_model=True,
    freqs=None,
):
    plt_dat = PlottingData()
    if pdgrm is not None:
        plt_dat.pdgrm = np.array(pdgrm.power, dtype=np.float64) * yscalar
        plt_dat.freqs = np.array(pdgrm.freqs)

    if spline_model is not None:

        if weights is None:
            # just use the initial weights/0 weights
            ln_spline = spline_model(use_parametric_model=use_parametric_model)

        elif weights.ndim == 1:
            # only one set of weights -- no CI possible
            ln_spline = spline_model(weights, use_parametric_model)

        else:  # weights.ndim == 2
            # multiple sets of weights -- CI possible
            ln_splines = jnp.array([spline_model(w, use_parametric_model) for w in weights])

            if use_uniform_ci:
                ln_ci = _get_uni_ci(ln_splines)
            else:  # percentile
                ln_ci = jnp.percentile(
                    ln_splines, q=jnp.array([16, 50, 84]), axis=0
                )
            ln_ci = jnp.array(ln_ci)
            plt_dat.ci = np.exp(ln_ci, dtype=np.float64) * yscalar
            ln_spline = ln_ci[1]
        plt_dat.model = np.exp(ln_spline, dtype=np.float64) * yscalar

    if plt_dat.freqs is None and freqs is None:
        plt_dat.freqs = np.linspace(0, 1, plt_dat.n)
    elif freqs is not None:
        plt_dat.freqs = freqs

    return plt_dat


def _get_uni_ci(samples, alpha=0.1):
    """
    Compute a uniform (simultaneous) confidence band for a set of function samples.

    Args:
        samples (jnp.ndarray): Shape (num_samples, num_points) array of function samples.
        alpha (float): Significance level (default 0.1 for 90% CI).

    Returns:
        tuple: (lower_bound, median, upper_bound) arrays.
    """
    num_samples, num_points = samples.shape

    # Compute pointwise median and standard deviation
    median = jnp.median(samples, axis=0)
    std = jnp.std(samples, axis=0)

    # Compute the max deviation over all samples
    deviations = (samples - median[None, :]) / std[
        None, :
    ]  # Normalize deviations
    max_deviation = jnp.max(
        jnp.abs(deviations), axis=1
    )  # Max deviation per sample

    # Compute the scaling factor using the distribution of max deviations
    k_alpha = jnp.percentile(
        max_deviation, 100 * (1 - alpha)
    )  # Critical value

    # Compute uniform confidence bands
    lower_bound = median - k_alpha * std
    upper_bound = median + k_alpha * std

    return lower_bound, median, upper_bound

1	import dataclasses	2✔
2
3	import arviz as az	2✔
4	import jax.numpy as jnp	2✔
5	import numpy as np	2✔
6
7	from ..datatypes import Periodogram	2✔
8	from ..psplines import LogPSplines	2✔
9
10	__all__ = ["unpack_data"]	2✔
11
12
13	@dataclasses.dataclass	2✔
14	class PlottingData:	2✔
15	freqs: np.ndarray = None	2✔
16	pdgrm: np.ndarray = None	2✔
17	model: np.ndarray = None	2✔
18	ci: np.ndarray = None	2✔
19
20	@property	2✔
21	def n(self):	2✔
22	if self.freqs is not None:	×
23	return len(self.freqs)	×
24	elif self.pdgrm is not None:	×
25	return len(self.pdgrm)	×
26	elif self.model is not None:	×
27	return len(self.model)	×
28	else:
29	raise ValueError("No data to get length from.")	×
30
31
32	def unpack_data(	2✔
33	pdgrm: Periodogram = None,
34	spline_model: LogPSplines = None,
35	weights=None,
36	yscalar=1.0,
37	use_uniform_ci=True,
38	use_parametric_model=True,
39	freqs=None,
40	):
41	plt_dat = PlottingData()	2✔
42	if pdgrm is not None:	2✔
43	plt_dat.pdgrm = np.array(pdgrm.power, dtype=np.float64) * yscalar	2✔
44	plt_dat.freqs = np.array(pdgrm.freqs)	2✔
45
46	if spline_model is not None:	2✔
47
48	if weights is None:	2✔
49	# just use the initial weights/0 weights
50	ln_spline = spline_model(use_parametric_model=use_parametric_model)	2✔
51
52	elif weights.ndim == 1:	2✔
53	# only one set of weights -- no CI possible
NEW 54	ln_spline = spline_model(weights, use_parametric_model)	×
55
56	else: # weights.ndim == 2
57	# multiple sets of weights -- CI possible
58	ln_splines = jnp.array([spline_model(w, use_parametric_model) for w in weights])	2✔
59
60	if use_uniform_ci:
61	ln_ci = _get_uni_ci(ln_splines)
62	else: # percentile	2✔
63	ln_ci = jnp.percentile(	2✔
64	ln_splines, q=jnp.array([16, 50, 84]), axis=0
UNCOV 65	)	×
66	ln_ci = jnp.array(ln_ci)
67	plt_dat.ci = np.exp(ln_ci, dtype=np.float64) * yscalar
68	ln_spline = ln_ci[1]	2✔
69	plt_dat.model = np.exp(ln_spline, dtype=np.float64) * yscalar	2✔
70		2✔
71	if plt_dat.freqs is None and freqs is None:	2✔
72	plt_dat.freqs = np.linspace(0, 1, plt_dat.n)
73	elif freqs is not None:	2✔
74	plt_dat.freqs = freqs	×
75		2✔
UNCOV 76	return plt_dat	×
77
78		2✔
79	def _get_uni_ci(samples, alpha=0.1):
80	"""
81	Compute a uniform (simultaneous) confidence band for a set of function samples.	2✔
82
83	Args:
84	samples (jnp.ndarray): Shape (num_samples, num_points) array of function samples.
85	alpha (float): Significance level (default 0.1 for 90% CI).
86
87	Returns:
88	tuple: (lower_bound, median, upper_bound) arrays.
89	"""
90	num_samples, num_points = samples.shape
91
92	# Compute pointwise median and standard deviation	2✔
93	median = jnp.median(samples, axis=0)
94	std = jnp.std(samples, axis=0)
95		2✔
96	# Compute the max deviation over all samples	2✔
97	deviations = (samples - median[None, :]) / std[
98	None, :
99	] # Normalize deviations	2✔
100	max_deviation = jnp.max(
101	jnp.abs(deviations), axis=1
102	) # Max deviation per sample	2✔
103
104	# Compute the scaling factor using the distribution of max deviations
105	k_alpha = jnp.percentile(
106	max_deviation, 100 * (1 - alpha)
107	) # Critical value	2✔
108
109	# Compute uniform confidence bands
110	lower_bound = median - k_alpha * std
111	upper_bound = median + k_alpha * std
112		2✔
113	return lower_bound, median, upper_bound	2✔

nz-gravity / LogPSplinePSD / 16343364439

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous