U
|
¬ý°dÝt ã @ sê d dl mZ d dlZd dlmZmZ d dlZd dlZd dl m
|
m Z d dl mZ d dlmZ d dlmZ d dlmZ d dlmZ d d lmZmZmZmZm Z d d
|
l m Z d d l m Z d dl!m" m#Z$ d d l%m&Z& d d l'm(Z( d dl)m*Z* d dl+m,Z,m-Z- d dl.m/Z/ d dl0m1Z1m2Z2m3Z3m4Z4m5Z5 erBd dl6m7Z7 d dl8m9Z9 G dd dZ:d8ddZ;d9ddddZ<d:d dd d Z=d d d d!d"Z>d;d d%d&d'd(Z?d<d%d)d*d+Z@d,d,d-d.d/ZAd=d d0d%d d1d2d3ZBd4d0d0d4d5d6d7ZCdS )>é )Ú annotationsN)Ú TYPE_CHECKINGÚcast)Únpt)ÚPerformanceWarning)Úcache_readonly)Úfind_stack_level)Ú maybe_promote)Úensure_platform_intÚis_1d_only_ea_dtypeÚis_extension_array_dtypeÚ
|
is_integerÚneeds_i8_conversion)ÚExtensionDtype)Únotna)Úfactorize_from_iterable)Ú ensure_wrapped_if_datetimelike)Ú DataFrame)ÚIndexÚ
|
MultiIndex)ÚSeries)Úcompress_group_indexÚdecons_obs_group_idsÚget_compressed_idsÚget_group_indexÚget_group_index_sorter)ÚExtensionArray)Ú
|
FrozenListc @ sÌ e Zd ZdZd)dddddZed d
|
d d Zed d
|
ddZdddddZdd Z edd
|
ddZ
|
edd
|
ddZ d d
|
d d Z d*d d Z d!d"d#d$Zedd
|
d%d&Zedd
|
d'd(ZdS )+Ú
|
_Unstackera
|
Helper class to unstack data / pivot with multi-level index
|
|
Parameters
|
----------
|
index : MultiIndex
|
level : int or str, default last level
|
Level to "unstack". Accepts a name for the level.
|
fill_value : scalar, optional
|
Default value to fill in missing values if subgroups do not have the
|
same set of labels. By default, missing values will be replaced with
|
the default fill value for that data type, NaN for float, NaT for
|
datetimelike, etc. For integer types, by default data will converted to
|
float and missing values will be set to NaN.
|
constructor : object
|
Pandas ``DataFrame`` or subclass used to create unstacked
|
response. If None, DataFrame will be used.
|
|
Examples
|
--------
|
>>> index = pd.MultiIndex.from_tuples([('one', 'a'), ('one', 'b'),
|
... ('two', 'a'), ('two', 'b')])
|
>>> s = pd.Series(np.arange(1, 5, dtype=np.int64), index=index)
|
>>> s
|
one a 1
|
b 2
|
two a 3
|
b 4
|
dtype: int64
|
|
>>> s.unstack(level=-1)
|
a b
|
one 1 2
|
two 3 4
|
|
>>> s.unstack(level=0)
|
one two
|
a 1 3
|
b 2 4
|
|
Returns
|
-------
|
unstacked : DataFrame
|
éÿÿÿÿNr ÚNone)ÚindexÚreturnc C sî |d kr t }|| _| ¡ | _| j |¡| _d| jj| j kr@dnd| _t| jj | _
|
t| jj | _ | j | j¡| _| j
|
| j¡| _|j | j | _t dd | j
|
D ¡}| jj}|| }|t tj¡jkrâtjd| dtt d | ¡ d S ) Nr é r c S s g | ]
|
}|j qS © )Úsize)Ú.0Z index_levelr$ r$ úRd:\z\workplace\vscode\pyvenv\venv\Lib\site-packages\pandas/core/reshape/reshape.pyÚ
|
<listcomp>y s z'_Unstacker.__init__.<locals>.<listcomp>z%The following operation may generate z& cells in the resulting pandas object.)Ú
|
stacklevel)r Ú constructorZremove_unused_levelsr! Ú_get_level_numberÚlevelÚcodesÚliftÚlistÚlevelsÚnew_index_levelsÚnamesÚnew_index_namesÚpopÚ removed_nameÚ removed_levelÚremoved_level_fullÚnpÚmaxr% ZiinfoÚint32ÚwarningsÚwarnr r Ú_make_selectors)Úselfr! r, r* Znum_rowsZ num_columnsZ num_cellsr$ r$ r' Ú__init__a s*
|
|
üz_Unstacker.__init__z-tuple[npt.NDArray[np.intp], list[np.ndarray]])r" c
|
C s | j }t| jj}t| jj}|d |
||d d
|| g }tdd |d |
||d d
|| g D }t||\}}t|}t||} | |fS )Nr# c s s | ]}t |V qd S ©N©Úlen©r& Úxr$ r$ r' Ú <genexpr> s z2_Unstacker._indexer_and_to_sort.<locals>.<genexpr>) r, r/ r! r- r0 Útupler rB r )
|
r> Úvr- ÚlevsÚto_sortÚsizesÚ
|
comp_indexÚobs_idsÚngroupsÚindexerr$ r$ r' Ú_indexer_and_to_sort s &4
|
z _Unstacker._indexer_and_to_sortzlist[np.ndarray]c s | j \ } fdd|D S )Nc s g | ]}| ¡qS r$ ©Útake)r& Úline©rN r$ r' r( ¡ s z,_Unstacker.sorted_labels.<locals>.<listcomp>)rO )r> rI r$ rS r' Ú sorted_labels s
|
z_Unstacker.sorted_labelsz
|
np.ndarray)Úvaluesr" c C s | j \}}tj||dd}|S )Nr )Úaxis)rO ÚalgosZtake_nd)r> rU rN Ú_Ú sorted_valuesr$ r$ r' Ú_make_sorted_values£ s
|
z _Unstacker._make_sorted_valuesc
|
C sØ | j }| jd d
}tdd |D }t||\}}t|}t|}| jj| j | j }||f| _
|
| jd || | j }t j t | j
|
¡td} | |d¡ | ¡ t| jk r¶td|| _| | _| t |¡¡| _d S )Nr c s s | ]}t |V qd S r@ rA rC r$ r$ r' rE ® s z-_Unstacker._make_selectors.<locals>.<genexpr>©ÚdtypeTz0Index contains duplicate entries, cannot reshape)r1 rT rF r rB r
|
r! Zlevshaper, r. Ú
|
full_shaper8 ÚzerosÚprodÚboolÚputÚsumÚ
|
ValueErrorÚ group_indexÚmaskZ searchsortedÚarangeÚ
|
compressor)
|
r> Ú
|
new_levelsZremaining_labelsZ level_sizesrK rL rM ÚstrideÚselectorre r$ r$ r' r= © s
|
z_Unstacker._make_selectorsr` c C s t | j ¡ S r@ )r` re Úall©r> r$ r$ r' Úmask_all s z_Unstacker.mask_allz2tuple[npt.NDArray[np.intp], npt.NDArray[np.bool_]]c C s6 t jt| jt jd}| j|dd\}}|| d¡fS )Nr[ r ©Ú
|
fill_valuer )r8 rf rB r! ZintpÚget_new_valuesÚany)r> Z dummy_arrÚ
|
new_valuesre r$ r$ r' Ú arange_resultÆ s z_Unstacker.arange_resultr c C sn |j dkr |d d
tjf }|d kr:|jd dkr:td| ||¡\}}| |¡}| j}| j||||j dS )Nr# z-must pass column labels for multi-column data)r! Úcolumnsr\ )
|
Úndimr8 ÚnewaxisÚshaperc rp Úget_new_columnsÚ new_indexr* r\ )r> rU Ú value_columnsro rX rt r! r$ r$ r' Ú
|
get_resultÎ s
|
|
ÿz_Unstacker.get_resultc C s |j dkr |d d
tjf }| |¡}| j\}}|jd }|| }||f}| j} | j}
|
|
|
rt|r| |||¡
|
dd¡ |¡} tj |t d} | | fS |j } |
|
r´|j } tj|| d} nTt| trâ| ¡ }|j|| d} || d d
< n&t| |\} }tj|| d} | |¡ | j}tj|t d} t|j r>| d¡}| d¡} n|j|dd}t || d¡|||| | d¡¡ t|j r| d¡} t | } | |j ¡} | | fS ) Nr# é r[ Úi8F)ÚcopyÚu1zM8[ns]) ru r8 rv rZ r] rw re rm rB ÚreshapeZswapaxesZonesr` r\ ÚemptyÚ
|
isinstancer Úconstruct_array_typeÚ_emptyr ÚfillÚnamer^ r ÚviewZastypeÚ
|
libreshapeÚunstackr )r> rU ro rY ÚlengthÚwidthri Z result_widthZ result_shapere rm rr Znew_maskr\ Úclsr r$ r$ r' rp Ý sd
|
|
|
|
ÿþÿ
|
|
|
ù
|
z_Unstacker.get_new_valuesz Index | None)rz c sì |d krB| j dkr"| jj| jdS | jjd| jjd}| | j¡S t| j| j }t|}t t
|
|¡|¡ t |t r¨|j | jf }|j| jf } fdd|jD }n || jg}|j| jg} g}| j}| t ||¡¡ t |||ddS )Nr ©r )Úitemc s g | ]}| ¡qS r$ rP ©r& Zlab©Z
|
propagatorr$ r' r( A s z._Unstacker.get_new_columns.<locals>.<listcomp>F©r0 r- r2 Zverify_integrity)r. r6 Ú_renamer5 ÚinsertÚ _na_valueÚrenamerB r8 Úrepeatrf r r r0 r7 r2 r- r Ú _repeaterÚappendÚtile) r> rz Úlevri r rh Ú new_namesÚ new_codesÚrepeaterr$ r r' rx , s6
|
|
ÿþ ÿz_Unstacker.get_new_columnsc C s\ t | jt | jkr8| j | j¡}| jrXt |dd¡}n t | j| j }t |¡| j }|S )Nr r )rB r7 r6 Z get_indexerr. r8 r rf )r> r ri r$ r$ r' r R s z_Unstacker._repeaterc s fdd j d d
D }t jdkrr jd |d }}|dk ¡ r\| t||j¡}| |¡ jd ¡S t j| jddS )Nc s g | ]}| j¡qS r$ )rQ rg r rl r$ r' r( d s z(_Unstacker.new_index.<locals>.<listcomp>r r# r Fr )
|
rT rB r1 rq r r rQ r r3 r )r> Z result_codesr, Ú level_codesr$ rl r' ry a s üz_Unstacker.new_index)r N)N)Ú__name__Ú
|
__module__Ú __qualname__Ú__doc__r? r rO rT rZ r= rm rs r{ rp rx r ry r$ r$ r$ r' r 3 s& -)
|
O&r c sz t dkr| S | j jkr& g fdd D fddtjD }fdd D }fdd D }fdd D }fdd|D }fd d|D }fd
|
d|D } td d |D }
|
t||
|
d d d} t| d d\} } t| | |
|
|d d}|st | dd}n"t
|
|| g || g | dg d d}t | t rl| ¡ }||_|jd|d}|}|}|}nàt | jt
|
r¶| } r² d¡|j|d}fdd D q~|S | j d d}||_|jd|d}t |t rê|j}n|j}t |t
|
s t|jd g| }| jjg| }|jd g}|D ] }| | |jd ¡¡ q.t
|
|||d d}t |t rp||_n||_|S )Nr c s g | ]} |¡qS r$ )r+ ©r& Úi©r! r$ r' r( s z%_unstack_multiple.<locals>.<listcomp>c s g | ]}| kr|qS r$ r$ r£ )Úclocsr$ r' r( s c s g | ]} j | qS r$ ©r0 r£ r¥ r$ r' r(
s c s g | ]} j | qS r$ ©r- r£ r¥ r$ r' r( s c s g | ]} j | qS r$ ©r2 r£ r¥ r$ r' r( s c s g | ]} j | qS r$ r§ r£ r¥ r$ r' r( s c s g | ]} j | qS r$ r¨ r£ r¥ r$ r' r( s c s g | ]} j | qS r$ r© r£ r¥ r$ r' r( s c s s | ]}t |V qd S r@ rA rC r$ r$ r' rE s z$_unstack_multiple.<locals>.<genexpr>F)ÚsortÚxnull)rª )r« Ú__placeholder__r r rn c s g | ]}| k r|n|d qS ©r# r$ ©r& rG )Úvalr$ r' r( « s ©Údeepr )rB r! r2 ÚrangeÚnlevelsrF r r r r r r r r~ r rt r4 ÚAssertionErrorr0 r r- r rQ )Údatar¦ ro ZrlocsZclevelsZccodesZcnamesZrlevelsZrcodesZrnamesrw rd Zcomp_idsrL Z recons_codesZ dummy_indexÚdummyZ unstackedrh r r ÚresultZunstcolsZrecÚ new_columnsr$ )r¦ r! r¯ r' Ú_unstack_multipleu sz
|
ü
|
ÿ r¹ zSeries | DataFrame)Úobjc C sâ t |ttfr0t|dkr(t| ||dS |d }t|sL|dksL| j |¡ t | trt | jt rpt
|
| ||dS | j j ddS n^t | jt s¤t dt| j dn:t| jrºt| ||S t| j|| jd }|j| jd |d
|
S d S ) Nr# rn r r¬ F©Údropnaz'index must be a MultiIndex to unstack, z was passed©r, r* ©rz ro )r rF r/ rB r¹ r r! r+ r r Ú_unstack_frameÚTÚstackrc Útyper r\ Ú_unstack_extension_seriesr Z_constructor_expanddimr{ Ú_values)rº r, ro Ú unstackerr$ r$ r' r Ì s4
|
ÿ
|
ÿ ÿr r c C sZ t | jtstt| j|| jd}| jsB| jj||d}| |¡S |j | j
|
| j |dS d S )Nr½ rn r¾ ) r r! r r´ r Ú _constructorZ_can_fast_transposeZ_mgrr r{ rÄ rt )rº r, ro rÅ Zmgrr$ r$ r' r¿ ð s
|
ÿr¿ r )Úseriesr" c C s( | ¡ }|j||d}|j d¡|_|S )a'
|
Unstack an ExtensionArray-backed Series.
|
|
The ExtensionDtype is preserved.
|
|
Parameters
|
----------
|
series : Series
|
A Series with an ExtensionArray for values
|
level : Any
|
The level name or number.
|
fill_value : Any
|
The user-level (not physical storage) fill value to use for
|
missing values introduced by the reshape. Passed to
|
``series.values.take``.
|
|
Returns
|
-------
|
DataFrame
|
Each column of the DataFrame will have the same dtype as
|
the input Series.
|
)r, ro r )Zto_framer rt Z droplevel)rÇ r, ro Zdfr· r$ r$ r' rà ý s rà r Tr` )Úframer¼ c s² dd }| j \} | j |¡}t| jtr8t| ||dS t| jtrÀt| jj} fdd| jj D }|| j\}} |
|
|¡ |
|
t | |¡ ¡ ¡ t| jj}
|
|
|
|
| jj¡ t|||
|
dd} nTtt|| j| jf \} \} } | ¡t | |¡ ¡ f}t| || jj| jjgdd} | js|| jr|t| jj}|d }t|rp| ¡ }| d d | ¡ D ¡}t|| }n
|
| j ¡ }n
|
| j ¡ }|r¤t |}|| }| | } | j || d
|
S ) zÀ
|
Convert DataFrame to Series with multi-level Index. Columns become the
|
second level of the resulting hierarchical index
|
|
Returns
|
-------
|
stacked : Series or DataFrame
|
c S s, | j r| t t| ¡fS t| \}}||fS r@ )Z is_uniquer8 rf rB r )r! r- Ú
|
categoriesr$ r$ r' Ú factorize( s zstack.<locals>.factorize)Ú level_numr¼ c s g | ]}| ¡qS r$ ©r r ©ÚKr$ r' r( 7 s zstack.<locals>.<listcomp>Fr r c S s g | ]\}}|j qS r$ ©rÄ )r& rX Úcolr$ r$ r' r( U s r¥ ) rw rt r+ r r Ú_stack_multi_columnsr! r/ r0 r- r r8 r Úravelr2 r ÚzipÚmapr r Ú_is_homogeneous_typeÚdtypesrÄ r r Ú_concat_same_typeÚitemsÚ"_reorder_for_extension_array_stackr Z_constructor_sliced)rÈ r, r¼ rÊ ÚNrË rh r ZclevZclabr ry r0 Zilabr- rÖ r\ Úarrrr re r$ rÍ r' rÁ sV
|
|
|
ÿ ü
|
ÿ
|
rÁ r» c s t fdd|D r4 }|D ]t||d}q nbt dd |D r } fdd|D }|r| d¡t||d}fdd|D }q\ntd |S )
|
Nc 3 s | ]}| j jkV qd S r@ )rt r2 ©r& r ©rÈ r$ r' rE k s z!stack_multiple.<locals>.<genexpr>r» c s s | ]}t |tV qd S r@ )r ÚintrÜ r$ r$ r' rE q s c s g | ]} j |¡qS r$ )rt r+ rÜ rÝ r$ r' r( w s z"stack_multiple.<locals>.<listcomp>r c s g | ]}| kr|n|d qS r r$ r® ©r r$ r' r( ~ s zTlevel should contain all level names or all level numbers, not a mixture of the two.)rk rÁ r4 rc )rÈ r, r¼ r· r$ )rÈ r r' Ústack_multipleh s
|
ÿrà r )rt r" c C s t | jdkr&| jd j| jd dS dd t| jdd
| jdd
D }t| }dd t |¡D }t| }tj d
|
d t|| jD | jdd
d S ) zBCreates a MultiIndex from the first N-1 levels of this MultiIndex.r| r r c s" g | ]\ } fd d|D qS )c s g | ]}|d kr | ndqS )r Nr$ )r& Úcrß r$ r' r( s z8_stack_multi_column_index.<locals>.<listcomp>.<listcomp>r$ )r& r- r$ rß r' r( s ÿz-_stack_multi_column_index.<locals>.<listcomp>Nr c s s | ]\}}|V qd S r@ r$ )r& ÚkeyrX r$ r$ r' rE s z,_stack_multi_column_index.<locals>.<genexpr>c S s* g | ]"\}}d |kr"t ||jdn|qS )Nr[ )r r\ )r& Znew_levr r$ r$ r' r( s ÿr© )
|
rB r0 r r2 rÓ r- Ú itertoolsÚgroupbyr Z from_arrays)rt rH ZtuplesZ unique_tuplesZnew_levsr$ r$ r' Ú_stack_multi_column_index s þ
|
ý úrå rÞ )rÈ rË r¼ r" c # sv ddddd}| j dd}|j}t|ts0t||jd kr|}t||jd D ](}|||}||d |} | || ¡}qR| |_}| ¡ s¬|d |}
|
|j |
|
dd
|
}|j}t
|
t|}t |} i } |j d } t t|jd }| t| d ¡}t ||¡}t| g }| D ]z}z|j |¡}W n& tk
|
rF | |¡ Y q
|
Y nX t|ts^t|}n |j|j }| kr°|jd d
|j| f }| |jjd ¡|_|j|d j }n¸| j r<t | j j d r<||j| j j d }||j| }| !¡ "d d | #¡ D ¡}|j$\}}t %|| ¡ &||¡j' (¡ }| |¡}n,| j)rV||j| j }n|j d d
|f }|j*dkr|| (¡ }|| |< q
|
t|d kr | +|¡} t|}t|j,træt-|j,j } t-|j,j.} fdd|j,jD } n*t/|j,\} } | g} | 0 ¡g} |j,j1g} | | ¡ | t 2||¡¡ | | jj.| ¡ t| | | dd}!| j3| |!| d}"|rr|"j4d dd}"|"S )NrÞ r ©rË rt c S s | |j kr|j | S | S )zì
|
Logic for converting the level number to something we can safely pass
|
to swaplevel.
|
|
If `level_num` matches a column name return the name from
|
position `level_num`, otherwise return `level_num`.
|
r© ræ r$ r$ r' Ú_convert_level_number§ s
|
|
z3_stack_multi_columns.<locals>._convert_level_numberFr° r# r )r, rV r )rt c S s g | ]\}}|j qS r$ rÏ )r& rX rD r$ r$ r' r( õ s z(_stack_multi_columns.<locals>.<listcomp>c s g | ]}| ¡qS r$ rÌ r ©Zlevsizer$ r' r( s r )r! rt rk )rV Úhow)5r~ rt r r r´ r³ r² Z swaplevelZ _is_lexsortedZ
|
sort_indexr rå r0 ÚsortedÚsetr- r rB r8 rQ Zget_locÚKeyErrorr ÚsliceÚstopÚstartÚlocZreindexrU rÕ r rÖ Zilocr r× rØ rw rf r rÀ rÒ Z_is_mixed_typeru Ú
|
differencer! r/ r2 r r r r rÆ r¼ )#rÈ rË r¼ rç ÚthisZmi_colsZ roll_columnsr¤ Zlev1Zlev2Z level_to_sortr¸ Znew_dataZ
|
level_valsr Zlevel_vals_nanZlevel_vals_usedZ drop_colsrâ rð Z slice_lenÚchunkZ value_slicer\ ZsubsetrÚ rÎ Úidxrh r r Z old_codesZ
|
old_levelsry r· r$ rè r' rÑ ¤ s
|
|
|
|
|
|
|
|
|
|
|
ÿÿ
|
|
|
|
ÿrÑ r )rÛ Ún_rowsÚ n_columnsr" c C s& t || ¡ ||¡j ¡ }| |¡S )aæ
|
Re-orders the values when stacking multiple extension-arrays.
|
|
The indirect stacking method used for EAs requires a followup
|
take to get the order correct.
|
|
Parameters
|
----------
|
arr : ExtensionArray
|
n_rows, n_columns : int
|
The number of rows and columns in the original DataFrame.
|
|
Returns
|
-------
|
taken : ExtensionArray
|
The original `arr` with elements re-ordered appropriately
|
|
Examples
|
--------
|
>>> arr = np.array(['a', 'b', 'c', 'd', 'e', 'f'])
|
>>> _reorder_for_extension_array_stack(arr, 2, 3)
|
array(['a', 'c', 'e', 'b', 'd', 'f'], dtype='<U1')
|
|
>>> _reorder_for_extension_array_stack(arr, 3, 2)
|
array(['a', 'd', 'b', 'e', 'c', 'f'], dtype='<U1')
|
)r8 rf r rÀ rÒ rQ )rÛ rõ rö rô r$ r$ r' rÙ ' s ! rÙ )N)N)N)r T)T)r T)DÚ
|
__future__r rã Útypingr r r; Únumpyr8 Zpandas._libs.reshapeZ_libsr r Zpandas._typingr Z pandas.errorsr Zpandas.util._decoratorsr Zpandas.util._exceptionsr Zpandas.core.dtypes.castr Zpandas.core.dtypes.commonr
|
r r r r Zpandas.core.dtypes.dtypesr Zpandas.core.dtypes.missingr Zpandas.core.algorithmsÚcoreZ
|
algorithmsrW Z pandas.core.arrays.categoricalr Zpandas.core.constructionr Zpandas.core.framer Zpandas.core.indexes.apir r Zpandas.core.seriesr Zpandas.core.sortingr r r r r Zpandas.core.arraysr Zpandas.core.indexes.frozenr r r¹ r r¿ rà rÁ rà rå rÑ rÙ r$ r$ r$ r' Ú<module> sL D
|
W$ !J! ÿ
|
