feat: add private _Analysis API to create analysis objects from standalone...

out/*

__pycache__/*

.venv/*

.devcontainer

from .analysis import (  # noqa:F401

    BaseAnalysis, CPAAnalysis, DPAAnalysis,

    ANOVAAnalysis, NICVAnalysis, SNRAnalysis,

    BasePartitionedAnalysis, MIAAnalysis

    BasePartitionedAnalysis, MIAAnalysis,

    _Analysis

)

from .preprocesses import preprocess, Preprocess, PreprocessError  # noqa:F401

from . import preprocesses  # noqa: F401

from . import selection_functions as _sf, container as _container, models, distinguishers

import inspect

import numpy as _np

import logging

    def __init__(self, bins_number=128, bin_edges=None, *args, **kwargs):

        distinguishers.mia._set_histogram_parameters(self, bins_number=bins_number, bin_edges=bin_edges)

        return super().__init__(*args, **kwargs)

class _Analysis:

    """Returns an analysis object created from a standalone distinguisher.

    It has been implemented for backward compatibility with previous eshard libraries and is not intented as a public API.

    """

    def __new__(cls, distinguisher, *args, **kwargs):

        klass = type.__new__(type, f'{distinguisher.__class__.__name__}Analysis', (BaseAnalysis, type(distinguisher).__bases__[1]), {})

        obj = klass(*args, **kwargs)

        init_args = inspect.getfullargspec(type(distinguisher)).args

        values = inspect.getmembers(distinguisher)

        for arg in init_args[1:]:

            val = list(filter(lambda t: t[0] == arg, values))[0]

            if val:

                setattr(obj, arg, val[1])

        return obj