U Qv fT9 @sdZddlZddlZddlZddlZddlZ ddl m Z ddl m Z dddd d d d d ddg Z d#ddZd$ddZddZddZd%ddZd&dd Zd'dd Zdd Zd(dd Zd)d dZd*d!d Zd+d"dZdS),zHelper functions.N)tabulate)options _perm_pval print_table_postprocess_dataframe _check_eftype remove_na _flatten_list_check_dataframe_is_sklearn_installed_is_statsmodels_installed_is_mpmath_installed two-sidedcCs|dkstdt|ttfs"tt|}|jdks>td|j}|dksTtd|dkrrt|| |}n<|dkrt || |}ntt |t | |}|S)ah Compute p-values from a permutation test. Parameters ---------- bootstat : 1D array Permutation distribution. estimate : float or int Point estimate. alternative : str Tail for p-value. Can be either `'two-sided'` (default), `'greater'` or `'less'`. Returns ------- p : float P-value. )rgreaterlesszWrong tail argument.rzbootstat must be a 1D array.z&bootstat must have at least one value.rr) AssertionError isinstanceintfloatnpasarrayndimsizeZ greater_equalsumZ less_equalfabsabs)ZbootstatZestimate alternativeZn_bootpr2/tmp/pip-unpacked-wheel-2te3nxqf/pingouin/utils.pyrs .3fsimplecCsJd|krtdd|kr(tdtt|dd||dtdd S) aPretty display of table. Parameters ---------- df : :py:class:`pandas.DataFrame` Dataframe to print (e.g. ANOVA summary) floatfmt : string Decimal number formatting tablefmt : string Table format (e.g. 'simple', 'plain', 'html', 'latex', 'grid', 'rst'). For a full list of available formats, please refer to https://pypi.org/project/tabulate/ Fz+ ============= ANOVA SUMMARY ============= Az. ============== POST HOC TESTS ============== keysF)headersZ showindexfloatfmttablefmtN)r%printr)dfr'r(rrr r>s   cCs|}t|j|jD]\}}t||}|dkr4qt|rt||j||f}||t |||<||j||f<qt |j||ft rqt |j||ft j }t |j||ftj}t||gsq|rt|j||fjj tj}|sqtj|j||f|d|j||f<q|S)a<Apply some post-processing to an ouput dataframe (e.g. rounding). Whether and how rounding is applied is governed by options specified in `pingouin.options`. The default rounding (number of decimals) is determined by `pingouin.options['round']`. You can specify rounding for a given column name by the option `'round.column.'`, e.g. `'round.column.CI95%'`. Analogously, `'round.row.'` also works (where `rowname`) refers to the pandas index), as well as `'round.cell.[]x[dnd}|j dkrRdn|}t ||} |j| |d}|j| |d}||fS) aRemove missing values along a given axis in one or more (paired) numpy arrays. Parameters ---------- x, y : 1D or 2D arrays Data. ``x`` and ``y`` must have the same number of dimensions. ``y`` can be None to only remove missing values in ``x``. paired : bool Indicates if the measurements are paired or not. axis : str Axis or axes along which missing values are removed. Can be 'rows' or 'columns'. This has no effect if ``x`` and ``y`` are one-dimensional arrays. Returns ------- x, y : np.ndarray Data without missing values Examples -------- Single 1D array >>> import numpy as np >>> from pingouin import remove_na >>> x = [6.4, 3.2, 4.5, np.nan] >>> remove_na(x) array([6.4, 3.2, 4.5]) With two paired 1D arrays >>> y = [2.3, np.nan, 5.2, 4.6] >>> remove_na(x, y, paired=True) (array([6.4, 4.5]), array([2.3, 5.2])) With two independent 2D arrays >>> x = np.array([[4, 2], [4, np.nan], [7, 6]]) >>> y = np.array([[6, np.nan], [3, 2], [2, 2]]) >>> x_no_nan, y_no_nan = remove_na(x, y, paired=False) )rAr0zaxis must be rows or columns.NrBrz#y cannot be an empty list or array.rFrA)rrrrJrrrstrrrrDr8rE logical_andrF) rGyZpairedrCZx_no_nanZy_no_nanrHZy_maskrIZbothrrr r s6+         cCst|tjjs|Sg}|D]h}t|tjj}|rxt|ttfsN|t|qt|trl|rl|t|q||q||qdd|D}|S)a|Flatten an arbitrarily nested list into a new list. This can be useful to select pandas DataFrame columns. From https://stackoverflow.com/a/16176969/10581531 Examples -------- >>> from pingouin.utils import _flatten_list >>> x = ['X1', ['M1', 'M2'], 'Y1', ['Y2']] >>> _flatten_list(x) ['X1', 'M1', 'M2', 'Y1', 'Y2'] >>> x = ['Xaa', 'Xbb', 'Xcc'] >>> _flatten_list(x) ['Xaa', 'Xbb', 'Xcc'] >>> x = ['Xaa', ('Xbb', 'Xcc'), (1, 2), (1)] >>> _flatten_list(x) ['Xaa', ('Xbb', 'Xcc'), (1, 2), 1] >>> _flatten_list(x, include_tuple=True) ['Xaa', 'Xbb', 'Xcc', 1, 2, 1] cSsg|]}|dk r|qSNr).0rrrr =sz!_flatten_list..) r collectionsabcIterablerKtupleextendr append)rGZ include_tupleresultelZ el_is_iterrrr r s  cCs|dkrdSdSdS)zCheck validity of eftype) noneZhedgesZcohenrPZpointbiserialrz eta-squarez odds-ratioZaucZclesTFN)lower)Zeftyperrr rAs c CsLt|tjsr|jjdkrjzddlm}m}Wntk rHt dYqrXt||r`| |}qrt dnt dt dd||fDrt d||j j d krt d |d krt d |d krt|tttfst d|d kr|dkrt d|dkrt|tttfst d|dkrH||fD] }t|tttfs&t dq&|S)zChecks whether data is a dataframe or can be converted to a dataframe. If successful, a dataframe is returned. If not successful, a ValueError is raised. DataMatrixr)r\convertzGFailed to convert object to pandas dataframe (DataMatrix not available)z5Data must be a pandas dataframe or compatible object.css|]}|dkVqdSrNr)rOvrrr lsz#_check_dataframe..zDV and data must be specifiedfizDV must be numeric.)withinbetween interactionrEz2Effects must be: within, between, interaction, allraz'within must be a string, int or a list.Nz-subject must be specified when effects=withinrbz(between must be a string, int or a list.rcz=within and between must be specified when effects=interaction)rpdZ DataFrame __class____name__Z datamatrixr\r] ImportError ValueErrorZ to_pandasr8r:kindrKrlist) dataZdvrbrasubjectZeffectsr\Zcnvinputrrr r Ss:         cCs>zddl}d}Wntk r(d}YnX|r:|s:td|S)z"Check if statsmodels is installed.rNTFzHstatsmodels needs to be installed. Please use `pip install statsmodels`.) statsmodelsOSError) raise_errorrn is_installedrrr r s cCs>zddl}d}Wntk r(d}YnX|r:|s:td|S)zCheck if sklearn is installed.rNTFzEsklearn needs to be installed. Please use `pip install scikit-learn`.)sklearnro)rprrrqrrr r s cCs>zddl}d}Wntk r(d}YnX|r:|s:td|S)zCheck if mpmath is installed.rNTFz>mpmath needs to be installed. Please use `pip install mpmath`.)mpmathro)rprsrqrrr rs )r)r!r")rA)NFrA)F)NNNNNN)F)F)F)__doc__r6ZnumpyrZpandasrd itertoolsr-collections.abcrRrconfigr__all__rrrr1rJr r rr r r rrrrr s<   & =  X 1 7