E g.ndZddlmZddlmZmZddlmZmZm Z ddl Z ddl m Z ddlmZmZmZddlmZdd lmZmZdd lmZmZmZdd lmZdd lmZdd lm Z erddl!m"Z"m#Z#ddl$m%Z%m&Z&d0dZ'GddeZ(Gdde(Z)Gdde(Z*d1d"Z+d2d&Z,d3d)Z-d3d*Z.d4d+Z/d5d-Z0d6d/Z1dS)7z Module responsible for execution of NDFrame.describe() method. Method NDFrame.describe() delegates actual execution to function describe_ndframe(). ) annotations)ABCabstractmethod) TYPE_CHECKINGCallablecastN) Timestamp)DtypeObjNDFrameTnpt)validate_percentile) is_bool_dtypeis_numeric_dtype) ArrowDtypeDatetimeTZDtypeExtensionDtype) Float64Dtype)concat)format_percentiles)HashableSequence) DataFrameSeriesobjr includestr | Sequence[str] | Noneexclude percentiles#Sequence[float] | np.ndarray | Nonereturnc t|}|jdkrttd|}n t td|||}||}tt |S)a Describe series or dataframe. Called from pandas.core.generic.NDFrame.describe() Parameters ---------- obj: DataFrame or Series Either dataframe or series to be described. include : 'all', list-like of dtypes or None (default), optional A white list of data types to include in the result. Ignored for ``Series``. exclude : list-like of dtypes or None (default), optional, A black list of data types to omit from the result. Ignored for ``Series``. percentiles : list-like of numbers, optional The percentiles to include in the output. All should fall between 0 and 1. The default is ``[.25, .5, .75]``, which returns the 25th, 50th, and 75th percentiles. Returns ------- Dataframe or series description. rrr)rrr)r)_refine_percentilesndimSeriesDescriberrDataFrameDescriberdescriber )rrrr describerresults Q/var/www/sysmax/venv/lib/python3.11/site-packages/pandas/core/methods/describe.pydescribe_ndframer,7s8&k22K x1}}#Xs##   '[#&&     K  8 8F & ! !!c2eZdZdZd dZed d Zd S) NDFrameDescriberAbstractzAbstract class for describing dataframe or series. Parameters ---------- obj : Series or DataFrame Object to be described. rDataFrame | Seriesr Nonec||_dSNr#)selfrs r+__init__z!NDFrameDescriberAbstract.__init__ns r-rSequence[float] | np.ndarraycdS)zDo describe either series or dataframe. Parameters ---------- percentiles : list-like of numbers The percentiles to include in the output. N)r4rs r+r(z!NDFrameDescriberAbstract.describeqsr-N)rr0r r1)rr6r r0)__name__ __module__ __qualname____doc__r5rr(r8r-r+r/r/esR   ^   r-r/c&eZdZUdZded<d dZdS) r&z2Class responsible for creating series description.rrrr6r cLt|j}||j|Sr3)select_describe_funcr)r4r describe_funcs r+r(zSeriesDescriber.describes,, H  }TX{333r-N)rr6r r)r9r:r;r<__annotations__r(r8r-r+r&r&|s9<<KKK444444r-r&c@eZdZUdZded<dfd Zdd Zdd ZxZS)r'abClass responsible for creating dataobj description. Parameters ---------- obj : DataFrame DataFrame to be described. include : 'all', list-like of dtypes or None A white list of data types to include in the result. exclude : list-like of dtypes or None A black list of data types to omit from the result. rrrrrr r1c||_||_|jdkr|jjdkrt dt |dS)Nrz+Cannot describe a DataFrame without columns)rrr%columnssize ValueErrorsuperr5)r4rrr __class__s r+r5zDataFrameDescriber.__init__sZ  8q==S[-22JKK K r-rr6cZ|}g}|D]3\}}t|}||||4t |t fd|Ddd}|j|_|S)Nc>g|]}|dS)F)copy)reindex).0x col_namess r+ z/DataFrameDescriber.describe..s) = = =!QYYyuY - - = = =r-r"F)axissort) _select_dataitemsr?appendreorder_columnsrrErL) r4rdataldesc_seriesr@drPs @r+r(zDataFrameDescriber.describes  ""  = =IAv088M LLv{;; < < < <#E**  = = = =u = = =    L%%'' r-cV|jP|jItjdg}|j|}t |jdkr|j}nQ|jdkr |jd}t||j}n&|j|j|j}|S)zSelect columns to be described.Ndatetime)rrallz*exclude must be None when include is 'all')rr) rrnpnumberr select_dtypeslenrErG)r4default_includerXmsgs r+rTzDataFrameDescriber._select_datas L t|';46Iz3JO8))/)BBD4<  A%%x \U " "|'B oo%8DD8))  *D r-)rrrrrrr r1)rr6r r)r r) r9r:r;r<rAr5r(rT __classcell__)rIs@r+r'r's|  NNN      "r-r'rYSequence[Series]list[Hashable]cg}t}td|Dt}|D]5}|D]0}||vr*||||16|S)z,Set a convenient order for rows for display.c3$K|] }|jV dSr3)index)rNrOs r+ z"reorder_columns..s$33AG333333r-)key)setsortedrcaddrV)rYnames seen_names ldesc_indexesidxnamesnames r+rWrWsE #J33U333===M!## # #D:%%t$$$ T""" # Lr-r[rSequence[float]cddlm}t|}gd|zdgz}||||g||z| gz}t|j trbt|j tr9|j jdkrd}nZddl}t|}n4t#}n%|j jdvrt%j d}nd}||||j| S) zDescribe series containing numerical data. Parameters ---------- series : Series Series to be described. percentiles : list-like of numbers The percentiles to include in the output. rr)countmeanstdminmaxmNiufbfloatrkrudtype)pandasrrryrzr{r|quantiletolistr} isinstancerrrkindpyarrowfloat64rr`ru)r[rrformatted_percentiles stat_indexr\rpas r+describe_numeric_1drsO.{;;0003HHE7RJ  fjjllC //+ & & - - / / 0 ::<<. &,// flJ / / #| C''$$$$"2::<<00 NNEE  f $ $!! 6!:FKu E E EEr-rXpercentiles_ignoredcBgd}|}t||dk}|dkr|jd|jd}}d}ntjtj}}d}||||g}ddlm} | |||j |S)zDescribe series containing categorical data. Parameters ---------- data : Series Series to be described. percentiles_ignored : list-like of numbers Ignored, but in place to unify interface. )ryuniquetopfreqrNobjectrxr) value_countsrcrkilocr`nanryrrru) rXrrq objcounts count_uniquerrrr*rs r+describe_categorical_1drs / . .E!!##Iya011LaOA& q(9TFBFTjjllL#t 4F 6&DIU C C CCr-cddg}|}t||dk}||g}d}|dkr|jd|jd}}|jj} |j d} t|}|j | | | }n| | }|gdz }|||t| | t| | gz }n$|dd gz }|t jt jgz }d }dd lm} | |||j| S) zDescribe series containing timestamp data treated as categorical. Parameters ---------- data : Series Series to be described. percentiles_ignored : list-like of numbers Ignored, but in place to unify interface. ryrrNi8)rrfirstlast)tzrrrrxr)rrcryrkrdtrdropnavaluesviewr tzinfo tz_convert tz_localizer|r}r`rrrru) rXrrqrrr*rrrrasintrs r+$describe_timestamp_as_categorical_1dr"s}h E!!##Iya011L'+zz||\&BF EaOA& q(9T WZ $))$//nn : !bn..$$CC//"%%C 1111   eiikkb ) ) ) eiikkb ) ) )    % 2626"" 6&DIU C C CCr-cZddlm}t|}gd|zdgz}|||g||z|gz}||||j S)zDescribe series containing datetime64 dtype. Parameters ---------- data : Series Series to be described. percentiles : list-like of numbers The percentiles to include in the output. rrx)ryrzr|r})rkru) rrrryrzr|rrr}ru)rXrrrrr\s r+describe_timestamp_1drRs.{;;))),AAUGKJ tyy{{DHHJJ/ -- $ $ + + - - . 88::,  6!:DI 6 6 66r-rct|jrtSt|rtS|jjdkst |jtrtS|jjdkrtStS)zSelect proper function for describing series based on data type. Parameters ---------- data : Series Series to be described. Mr~) rrrrrrrrr)rXs r+r?r?jssTZ  '&& $  '"" C  :dj/#J#J $$ C  ""&&r-npt.NDArray[np.float64]cZ|tjgdSt|}t|d|vr|dtj|}tj|}|Jt|t|krtd|S)z Ensure that percentiles are unique and sorted. Parameters ---------- percentiles : list-like of numbers, optional The percentiles to include in the output. N)g??g?rz%percentiles cannot contain duplicates) r`arraylistr rVasarrayrrcrG)r unique_pctss r+r$r$sx)))***{##K $$$ +3*[))K)K((K  " " " ;#k****@AAA r-) rr rrrrrrr r )rYrgr rh)r[rrrvr r)rXrrrvr r)rXrrrvr r)rXrr r)rrr r)2r< __future__rabcrrtypingrrrnumpyr`pandas._libs.tslibsr pandas._typingr r r pandas.util._validatorsr pandas.core.dtypes.commonrrpandas.core.dtypes.dtypesrrrpandas.core.arrays.floatingrpandas.core.reshape.concatrpandas.io.formats.formatrcollections.abcrrrrrr,r/r&r'rWrrrrr?r$r8r-r+rs #""""" )))))) 877777 544444------777777   +"+"+"+"\     s   . 4 4 4 4 4. 4 4 4AAAAA1AAAH    &F&F&F&FRDDDD@-D-D-D-D`77770'''',      r-