>hk SrSSKJrJr SSKJr SSKrSSKJrJ r J r J r SSK J r SSKJr S/rS rSS jrS rS rSrSrSSjrSrSrSrSrSrSrSrSSS SSSSSS S4 Sjrg)zCreate a mosaic plot from a contingency table. It allows to visualize multivariate categorical data in a rigorous and informative way. see the docstring of the mosaic function for more informations. )lrangelzip)productN)arraycumsumiterabler_) DataFrame)utilsmosaiccX[U5(dXUS:Xa[SS/5nODUS:a[SS/5nO0US:a[SRU55e[USU- /5n[R "U[ S9n[R"US:5(a[SRU55e[R"US5(a[SRU55e[U5S:a [SS/5$[S[U54nXS S--nU$) z return a list of proportions of the available space given the division if only a number is given, it will assume a split in two pieces r?z.proportions should be positive,given value: {})dtypezBat least one proportion should be greater than zerogiven value: {}) rr ValueErrorformatnpasarrayfloatanyallcloselenr r) proportionlefts rC:\Users\julio\OneDrive\Documentos\Trabajo\Ideas Frescas\venv\Lib\site-packages\statsmodels/graphics/mosaicplot.py_normalize_splitrs( J   ?Sz*J 1_Sz*J !^006z0BD D C*,<=>JJe4J vvj1n,,2F:,>@ @ {{:q!! $fZ0   :c3Z  a ## $DHsND KTc [U5[U5[U5[U54uppxUS:dUS:a[SRXx55e[U5n U SSn U SSU - n X[R "[ U 5S- 5-- n U SU S-U S- n X-n X-n U(aUOUX(aUOU--n X(aUOU-n [X5V Vs/sHupU(aXX4OX X~4PM nn nU$s snn f)aQ Split the given rectangle in n segments whose proportion is specified along the given axis if a gap is inserted, they will be separated by a certain amount of space, retaining the relative proportion between them a gap of 1 correspond to a plot that is half void and the remaining half space is proportionally divided among the pieces. rz/dimension of the square less thanzero w={} h={}Nrr)rrrrrarangerzip)xywidthheightr horizontalgapwh proportionsstarting amplitude extensionsaresultss r _split_rectr37s0q58U5\5=@JA! A1q5++16!<9 9":.K3BHAB(*I biiK 01 4555H y},x{:I H I Q(:a1*MMH*Q!4I 464DA *a|a|;4 6 N6s Dch^^[T5m[UU4SjUR555nU$)zn Make partial sum on a counter dict. Given a match for the beginning of the category, it will sum each value. c3B># UHupUSTT:XdMUv M g7fN).0kvL partial_keys r _reduce_dict..ds$I0daAbqE[4H0s )rsumitems) count_dictr<countr;s ` @r _reduce_dictrC^s- KA Ij..0I IE Lr c 0n[U5nUR5HBunuppX8SU:Xa+[XXX$U5n [X5H upXX4-'M M<XX4Xh'MD U$)z Given a dictionary where each entry is a rectangle, a list of key and value (count of elements in each category) it split each rect accordingly, as long as the key start with the tuple key_subset. The other keys are returned without modification. N)rr@r3r#) rect_dictkeysvalues key_subsetr(r)resultr;namer$r%r*r+ divisionskeyrects r_key_splittingrNhsyF JA'oo/lqQ bq !#A!CHI 1 (,tf}%2!.s(CqCFFCs)rr isinstancerQtuple)objress r_tuplifyrX|sH {{3 3 4 4(C(( J3xk Jr c /n[U6H9n[U5nUR[UVs0sHoDS_M sn55 M; U$s snf)zuse the Ordered dict to implement a simple ordered set return each level of each category [[key_1_level_1,key_2_level_1],[key_1_level_2,key_2_level_2]] N)r#rXappendlist)rFrWi tuplefiedjs r_categories_levelr_sN C 4\QK  4)4)QD)456 J5s A c [[5S4/5n[[UR 555n[ U5n[ R"U5(d [U5Vs/sH obSU-- PM nn[ U5U:a/USn[U6[U5Vs/sH ogSU-- PM sn-nUSUn[[U65Vs0sHnXU_M n n[U5HZup[[USU 65n U H4n U Vs/sHn[XU4-5PM nnX*n[X;XUU5nM6 U(+nM\ U$s snfs snfs snfs snf)a Split a square in a hierarchical way given a contingency table. Hierarchically split the unit square in alternate directions in proportion to the subdivision contained in the contingency table count_dict. This is the function that actually perform the tiling for the creation of the mosaic plot. If the gap array has been specified it will insert a corresponding amount of space (proportional to the unit length), while retaining the proportionality of the tiles. Parameters ---------- count_dict : dict Dictionary containing the contingency table. Each category should contain a non-negative number with a tuple as index. It expects that all the combination of keys to be represents; if that is not true, will automatically consider the missing values as 0 horizontal : bool The starting direction of the split (by default along the horizontal axis) gap : float or array of floats The list of gaps to be applied on each subdivision. If the length of the given array is less of the number of subcategories (or if it's a single number) it will extend it with exponentially decreasing gaps Returns ------- base_rect : dict A dictionary containing the result of the split. To each key is associated a 4-tuple of coordinates that are required to create the corresponding rectangle: 0 - x position of the lower left corner 1 - y position of the lower left corner 2 - width of the rectangle 3 - height of the rectangle )rrrrg?rN) dictrUr_r[rFrrrranger enumeraterCrN)rAr(r) base_rectcategories_levelsr;idxlastr9 count_orderedcat_idxcat_enum base_keysrLpartial part_countnew_gaps r_hierarchical_splitrosRuw -./I)$z/@*AB A ;;s  +0848CSCZ84 3x!|2wCjqB#3#:-BB bq'C"&g/@&A!BD!BA1 %!BD&'89"3HW"=>? C,45+3'}WJ6FG+3 5lG&yJ'17.-s(CqCrc3.># UH nTUv M g7fr6r7)r8r\rLs rr=r>s.1As)hasattr_normalize_dataframer[r@AttributeErrorrrndindexshaperUrXr_rFrgetrr)rindexr@temprfrJr9r:reindexes contingencyrnew_keyrLs @r_normalize_datarsttW'$ ":":#D0 #TZZ\"(- -utqHQKNuD -)$tyy{*;<()G.56gdhhq!n$gK6 D/4mF3() *EKjjl U...$G# D K5  #zz$::djj)C(C((DDJ*TZZ\" # .7sD FF A.Zs/Aarrrrr@rr$rxrRry)r~r) rr_r[rFrr?rGrbr@rZrcrsqrt)rrertotal levels_count level_idxrrrLrexpectedbaser\r9mr0sigmaspropsdevredbluegreenrs r_statistical_coloringrSs 4 &D)$tyy{*;<#$G #/// /EL7^  &1E #J "jjl  N*%.%+   &  2 J'$Hjjl cNDA OA& &D# bggelcDj.I&JJ #5=NN4D E4Dyq&1aq'A+""4DF E ELLNS1Wc3!c'?AgsC28$4sT!S(Qw38C #UD1EBc # LFsGc2US:XaU$XS- -U-U-U- $)Nrrr7)r$r*r+Ws r _get_positionr~s)Av CK1 q 1 $$r c ^^^[[UR555m[T5S:a Sn[ U5e0n[UR 55nU(+nUR 5nUR5n URURU RUR/n URURU RUR/n U(aU SSU SS-n U SSU SS-n [X5HupU "/5 U "/5 M [T5GH:up[5nUHnU(aUS:Xa/SQmO/SQmOUS:Xa/SQmO/SQm[UU4S j[!U555nUU4-nUVVs0sHunnUUSUS-:XdMUU_M nnn[UR#55n[%S U55m[%U4S jU55n[%U4S jU55nX-S-nUS -(aUOUUU'M X"[UR#555 X"[UR55X>S9 GM= U$s snnf)zfind the position of the label for each value of each category right now it supports only up to the four categories ax: the axis on which the label should be applied rotation: the rotation list for each side zrmaximum of 4 level supported for axes labeling... and 4is already a lot of levels, are you sure you need them all?rNrr)rrr)rrr)rrrc3:># UHnTUTUv M g7fr6r7)r8r\ categories index_selects rr=!_create_labels..s%7%5'qM,q/:%5sc30# UH upp4X4-v M g7fr6r7)r8r$r%r*r+s rr=rs3dlqQAEdsc3D># UHupp4[XUT5v M g7fr6rr8r$r%r*r+rs rr=r!KdlqQ aAq11d c3D># UHupp4[X$UT5v M g7fr6rrs rr=rrrr)rotation)r_r[rFrrr@twinxtwiny set_xticks set_yticksset_xticklabelsset_yticklabelsr#rcrarUrbrGr?)rectsr(axrmsglabelsr@verticalax2ax3 ticks_pos ticks_labposlabrr level_ticksrbasekeyr9r:subsetvalsx_laby_labsiderrrs @@@r_create_labelsrs`#4 #56J :Mo F  E~H ((*C ((*C s~~s~~NI##R%7%7$$c&9&9;IabMIbqM1 abMIbqM1  - B B. &j1 f E >#/L#/L>#/L#/L7%*9%577G(G',;utq!$.9q=(99aduF; (D3d33AKdKKEKdKKE(A-D*.(K IN T+"4"4"678T+"2"2"45&.&9 ;W2Z M+;s I- (I- g{Gzt?cgr6r7)rLs rrs$r ryFrc P^[U[5(aUc [S5eSSKJn [ R "U5up[X5n[XUS9n UcSnU(a [U5nO [U5n[U[5(aUmU4SjnU R5HjunnUunnnnU"U5nU(aUOXnU"U5nU "UU4UU4SU0UD6nURU5 URUUS- -UUS- -US S S S 9 Ml U (a2[R "U 5(aU nOU /S -n[#XUU5nODUR%/5 UR'/5 UR)/5 UR+/5 UR-U5 X4$) aCreate a mosaic plot from a contingency table. It allows to visualize multivariate categorical data in a rigorous and informative way. Parameters ---------- data : {dict, Series, ndarray, DataFrame} The contingency table that contains the data. Each category should contain a non-negative number with a tuple as index. It expects that all the combination of keys to be represents; if that is not true, will automatically consider the missing values as 0. The order of the keys will be the same as the one of insertion. If a dict of a Series (or any other dict like object) is used, it will take the keys as labels. If a np.ndarray is provided, it will generate a simple numerical labels. index : list, optional Gives the preferred order for the category ordering. If not specified will default to the given order. It does not support named indexes for hierarchical Series. If a DataFrame is provided, it expects a list with the name of the columns. ax : Axes, optional The graph where display the mosaic. If not given, will create a new figure horizontal : bool, optional The starting direction of the split (by default along the horizontal axis) gap : {float, sequence[float]} The list of gaps to be applied on each subdivision. If the length of the given array is less of the number of subcategories (or if it's a single number) it will extend it with exponentially decreasing gaps properties : dict[str, callable], optional A function that for each tile in the mosaic take the key of the tile and returns the dictionary of properties of the generated Rectangle, like color, hatch or similar. A default properties set will be provided fot the keys whose color has not been defined, and will use color variation to help visually separates the various categories. It should return None to indicate that it should use the default property for the tile. A dictionary of the properties for each key can be passed, and it will be internally converted to the correct function labelizer : dict[str, callable], optional A function that generate the text to display at the center of each tile base on the key of that tile title : str, optional The title of the axis statistic : bool, optional If true will use a crude statistical model to give colors to the plot. If the tile has a constraint that is more than 2 standard deviation from the expected value under independence hypothesis, it will go from green to red (for positive deviations, blue otherwise) and will acquire an hatching when crosses the 3 sigma. axes_label : bool, optional Show the name of each value of each category on the axis (default) or hide them. label_rotation : {float, list[float]} The rotation of the axis label (if present). If a list is given each axis can have a different rotation Returns ------- fig : Figure The figure containing the plot. rects : dict A dictionary that has the same keys of the original dataset, that holds a reference to the coordinates of the tile and the Rectangle that represent it. References ---------- A Brief History of the Mosaic Display Michael Friendly, York University, Psychology Department Journal of Computational and Graphical Statistics, 2001 Mosaic Displays for Loglinear Models. Michael Friendly, York University, Psychology Department Proceedings of the Statistical Graphics Section, 1992, 61-68. Mosaic displays for multi-way contingency tables. Michael Friendly, York University, Psychology Department Journal of the american statistical association March 1994, Vol. 89, No. 425, Theory and Methods Examples -------- >>> import numpy as np >>> import pandas as pd >>> import matplotlib.pyplot as plt >>> from statsmodels.graphics.mosaicplot import mosaic The most simple use case is to take a dictionary and plot the result >>> data = {'a': 10, 'b': 15, 'c': 16} >>> mosaic(data, title='basic dictionary') >>> plt.show() A more useful example is given by a dictionary with multiple indices. In this case we use a wider gap to a better visual separation of the resulting plot >>> data = {('a', 'b'): 1, ('a', 'c'): 2, ('d', 'b'): 3, ('d', 'c'): 4} >>> mosaic(data, gap=0.05, title='complete dictionary') >>> plt.show() The same data can be given as a simple or hierarchical indexed Series >>> rand = np.random.random >>> from itertools import product >>> tuples = list(product(['bar', 'baz', 'foo', 'qux'], ['one', 'two'])) >>> index = pd.MultiIndex.from_tuples(tuples, names=['first', 'second']) >>> data = pd.Series(rand(8), index=index) >>> mosaic(data, title='hierarchical index series') >>> plt.show() The third accepted data structure is the np array, for which a very simple index will be created. >>> rand = np.random.random >>> data = 1+rand((2,2)) >>> mosaic(data, title='random non-labeled array') >>> plt.show() If you need to modify the labeling and the coloring you can give a function tocreate the labels and one with the graphical properties starting from the key tuple >>> data = {'a': 10, 'b': 15, 'c': 16} >>> props = lambda key: {'color': 'r' if 'a' in key else 'gray'} >>> labelizer = lambda k: {('a',): 'first', ('b',): 'second', ... ('c',): 'third'}[k] >>> mosaic(data, title='colored dictionary', properties=props, ... labelizer=labelizer) >>> plt.show() Using a DataFrame as source, specifying the name of the columns of interest >>> gender = ['male', 'male', 'male', 'female', 'female', 'female'] >>> pet = ['cat', 'dog', 'dog', 'cat', 'dog', 'cat'] >>> data = pd.DataFrame({'gender': gender, 'pet': pet}) >>> mosaic(data, ['pet', 'gender'], title='DataFrame as Source') >>> plt.show() .. plot :: plots/graphics_mosaicplot_mosaic.py z.}s diilr c(>TRUS5$r6)r)rL color_dicts rrrsT!:r labelrcentersmaller)havasizer)rTr rmatplotlib.patchesrr create_mpl_axrrorrrar@ add_patchtextrrrrrrr set_title)rrrr(r)r labelizertitle statistic axes_labellabel_rotationrfigr default_propsr9r:r$r%r*r+confrrRectrrrs @rr r sn$ ""u}*+ +-!!"%GC 4 'D  EE* -d3 248 *d## :  1 1a!}-"2|!QA;T;U; T AE 1q1u9dx9  . ;;~ & &%H&'!+H2x@ b 2 b 2LL :r )Tg?) __doc__statsmodels.compat.pythonrr itertoolsrnumpyrrrrr pandasr statsmodels.graphicsr __all__rr3rCrNrXr_rorvrrrrrrr r7r rrs3--& *@$N( IX> ,^-` (V% Pf5&$uFr