U 3dg,@sddlmZddlmZddlmZddlZddlm Z dddd d Z d!d d Z Gd ddeZ ddZ GdddeZddZddZddddZd"ddZGdddeZdd ZdS)#)suppress)Counter) NamedTupleN is_scalar_nanFreturn_inverse return_countscCs&|jtkrt|||dSt|||dS)aHelper function to find unique values with support for python objects. Uses pure python method for object dtype, and numpy method for all other dtypes. Parameters ---------- values : ndarray Values to check for unknowns. return_inverse : bool, default=False If True, also return the indices of the unique values. return_counts : bool, default=False If True, also return the number of times each unique item appears in values. Returns ------- unique : ndarray The sorted unique values. unique_inverse : ndarray The indices to reconstruct the original array from the unique array. Only provided if `return_inverse` is True. unique_counts : ndarray The number of times each of the unique values comes up in the original array. Only provided if `return_counts` is True. r)dtypeobject_unique_python _unique_np)valuesr r r9/tmp/pip-unpacked-wheel-zrfo1fqw/sklearn/utils/_encode.py_unique s rcCstj|||d}d\}}|r&|^}}|r4|^}}|s<|rD|d}|jrt|drt|tj}|d|d}|r||||k<|rt||d||<|d|d}|f}|r||f7}|r||f7}t|dkr|dS|S)zHelper function to find unique values for numpy arrays that correctly accounts for nans. See `_unique` documentation for details.r)NNrNr)npuniquesizer searchsortednansumlen)rr r uniquesZinversecountsZnan_idxretrrrr2s4     rc@s*eZdZUdZeed<eed<ddZdS) MissingValuesz'Data class for missing data informationrnonecCs*g}|jr|d|jr&|tj|S)z3Convert tuple to a list where None is always first.N)rappendrr)selfoutputrrrto_listas   zMissingValues.to_listN)__name__ __module__ __qualname____doc__bool__annotations__r#rrrrr[s rcCsndd|D}|s"|tdddfSd|krRt|dkrDtddd}q^tddd}n tddd}||}||fS)a.Extract missing values from `values`. Parameters ---------- values: set Set of values to extract missing from. Returns ------- output: set Set with missing values extracted. missing_values: MissingValues Object with missing value information. cSs h|]}|dkst|r|qSNr.0valuerrr {sz#_extract_missing..F)rrNrT)rr)rZmissing_values_setZoutput_missing_valuesr"rrr_extract_missingks  r/cs(eZdZdZfddZddZZS)_nandictz!Dictionary with support for nans.cs4t||D]\}}t|r||_q0qdSr*)super__init__itemsr nan_value)r!mappingkeyr- __class__rrr2s  z_nandict.__init__cCs$t|drt|r|jSt|dS)Nr4)hasattrrr4KeyErrorr!r6rrr __missing__sz_nandict.__missing__)r$r%r&r'r2r< __classcell__rrr7rr0s r0cs.tddt|Dtfdd|DS)z,Map values based on its position in uniques.cSsi|]\}}||qSrr)r,ivalrrr sz#_map_to_integer..csg|] }|qSrrr,vtablerr sz#_map_to_integer..)r0 enumeraterarray)rrrrCr_map_to_integersrHcCsz>t|}t|\}}t|}||tj||jd}WnBtk rtddtdd|DD}td|YnX|f}|r|t ||f7}|r|t ||f7}t |dkr|dS|S)Nr css|] }|jVqdSr*)r&)r,trrr sz!_unique_python..css|]}t|VqdSr*)typerArrrrKszEEncoders require their input to be uniformly strings or numbers. Got rr) setr/sortedextendr#rrGr TypeErrorrH _get_countsr)rr r uniques_setZmissing_valuesrtypesrrrrr s"   r T) check_unknownc Cs|jjdkrPz t||WStk rL}ztdt|W5d}~XYqXn0|rtt||}|rttdt|t||SdS)aHelper function to encode values into [0, n_uniques - 1]. Uses pure python method for object dtype, and numpy method for all other dtypes. The numpy method has the limitation that the `uniques` need to be sorted. Importantly, this is not checked but assumed to already be the case. The calling method needs to ensure this for all non-object values. Parameters ---------- values : ndarray Values to encode. uniques : ndarray The unique values in `values`. If the dtype is not object, then `uniques` needs to be sorted. check_unknown : bool, default=True If True, check for values in `values` that are not in `unique` and raise an error. This is ignored for object dtype, and treated as True in this case. This parameter is useful for _BaseEncoder._transform() to avoid calling _check_unknown() twice. Returns ------- encoded : ndarray Encoded values OUSz%y contains previously unseen labels: N) r kindrHr: ValueErrorstr_check_unknownrr)rrrTediffrrr_encodes  & r\c s~d}|jjdkrt|}t|\}}t|t\|}|joLj }|joZj }fdd|r|sz|sz|rtfdd|D}ntjt |t d}t |}|r| d|r| tjnt |} tj| |dd }|r|jrt||}ntjt |t d}t|rdt|} | rd|jrZ|rZt|} d || <|| }t |}|rz||fS|S) a Helper function to check for unknowns in values to be encoded. Uses pure python method for object dtype, and numpy method for all other dtypes. Parameters ---------- values : array Values to check for unknowns. known_values : array Known values. Must be unique. return_mask : bool, default=False If True, return a mask of the same shape as `values` indicating the valid values. Returns ------- diff : list The unique values present in `values` and not in `know_values`. valid_mask : boolean array Additionally returned if ``return_mask=True``. NrUcs$|kp"jr|dkp"jo"t|Sr*)rrr)r-)missing_in_uniquesrRrris_validsz _check_unknown..is_validcsg|] }|qSrrr+)r^rrrEsz"_check_unknown..rITZ assume_uniquer)r rVrMr/rrrrGZonesrr(listr rZ setdiff1drZin1disnanany) rZ known_valuesZ return_maskZ valid_maskZ values_setZmissing_in_valuesr[Z nan_in_diffZ none_in_diff unique_valuesZ diff_is_nanis_nanr)r^r]rRrrYsH           rYcs0eZdZdZfddZddZddZZS) _NaNCounterz$Counter with support for nan values.cst||dSr*)r1r2_generate_items)r!r3r7rrr2Csz_NaNCounter.__init__ccs<|D]2}t|s|Vqt|ds(d|_|jd7_qdS)z>Generate items without nans. Stores the nan counts separately. nan_countrrN)rr9rg)r!r3itemrrrrfFs z_NaNCounter._generate_itemscCs$t|drt|r|jSt|dS)Nrg)r9rrgr:r;rrrr<Psz_NaNCounter.__missing__)r$r%r&r'r2rfr<r=rrr7rre@s  rec Cs|jjdkr^t|}tjt|tjd}t|D](\}}tt ||||<W5QRXq0|St |dd\}}tj ||dd}t |drt |drd|d<t |||} tj|tjd}|| ||<|S)zGet the count of each of the `uniques` in `values`. The counts will use the order passed in by `uniques`. For non-object dtypes, `uniques` is assumed to be sorted and `np.nan` is at the end. ZOUrIT)r r_r)r rVrerzerosrZint64rFrr:risinrarZ zeros_like) rrcounterr"r>rhrcrZuniques_in_valuesZunique_valid_indicesrrrrQVs   rQ)FF)F) contextlibr collectionsrtypingrZnumpyrrrrrr/dictr0rHr r\rYrerQrrrrs    ) )&* U