M/Ph-AjdZddlZddlmZddlmZddgZ dd Zd Z d Z ddddd ifdZ dZ dZ dS)zVariations on boxplots.N) gaussian_kde)utils violinplotbeanplotbothTc |in|}td|Ddkrd}t|tj|\}}t t t j|}|$t jt|dz}t j|t j |z } t j dt j| dgz| dd d z g} t||D]\} } t|| | | |||r||d|d t||||||S) a Make a violin plot of each dataset in the `data` sequence. A violin plot is a boxplot combined with a kernel density estimate of the probability density function per point. Parameters ---------- data : sequence[array_like] Data arrays, one array per value in `positions`. ax : AxesSubplot, optional If given, this subplot is used to plot in instead of a new figure being created. labels : list[str], optional Tick labels for the horizontal axis. If not given, integers ``1..len(data)`` are used. positions : array_like, optional Position array, used as the horizontal axis of the plot. If not given, spacing of the violins will be equidistant. side : {'both', 'left', 'right'}, optional How to plot the violin. Default is 'both'. The 'left', 'right' options can be used to create asymmetric violin plots. show_boxplot : bool, optional Whether or not to show normal box plots on top of the violins. Default is True. plot_opts : dict, optional A dictionary with plotting options. Any of the following can be provided, if not present in `plot_opts` the defaults will be used:: - 'violin_fc', MPL color. Fill color for violins. Default is 'y'. - 'violin_ec', MPL color. Edge color for violins. Default is 'k'. - 'violin_lw', scalar. Edge linewidth for violins. Default is 1. - 'violin_alpha', float. Transparancy of violins. Default is 0.5. - 'cutoff', bool. If True, limit violin range to data range. Default is False. - 'cutoff_val', scalar. Where to cut off violins if `cutoff` is True. Default is 1.5 standard deviations. - 'cutoff_type', {'std', 'abs'}. Whether cutoff value is absolute, or in standard deviations. Default is 'std'. - 'violin_width' : float. Relative width of violins. Max available space is 1, default is 0.8. - 'label_fontsize', MPL fontsize. Adjusts fontsize only if given. - 'label_rotation', scalar. Adjusts label rotation only if given. Specify in degrees. - 'bw_factor', Adjusts the scipy gaussian_kde kernel. default: None. Options for scalar or callable. Returns ------- Figure If `ax` is None, the created figure. Otherwise the figure to which `ax` is connected. See Also -------- beanplot : Bean plot, builds on `violinplot`. matplotlib.pyplot.boxplot : Standard boxplot. Notes ----- The appearance of violins can be customized with `plot_opts`. If customization of boxplot elements is required, set `show_boxplot` to False and plot it on top of the violins by calling the Matplotlib `boxplot` function directly. For example:: violinplot(data, ax=ax, show_boxplot=False) ax.boxplot(data, sym='cv', whis=2.5) It can happen that the axis labels or tick labels fall outside the plot area, especially with rotated labels on the horizontal axis. With Matplotlib 1.1 or higher, this can easily be fixed by calling ``ax.tight_layout()``. With older Matplotlib one has to use ``plt.rc`` or ``plt.rcParams`` to fix this, for example:: plt.rc('figure.subplot', bottom=0.25) violinplot(data, ax=ax) References ---------- J.L. Hintze and R.D. Nelson, "Violin Plots: A Box Plot-Density Trace Synergism", The American Statistician, Vol. 52, pp.181-84, 1998. Examples -------- We use the American National Election Survey 1996 dataset, which has Party Identification of respondents as independent variable and (among other data) age as dependent variable. >>> data = sm.datasets.anes96.load_pandas() >>> party_ID = np.arange(7) >>> labels = ["Strong Democrat", "Weak Democrat", "Independent-Democrat", ... "Independent-Indpendent", "Independent-Republican", ... "Weak Republican", "Strong Republican"] Group age by party ID, and create a violin plot with it: >>> plt.rcParams['figure.subplot.bottom'] = 0.23 # keep labels visible >>> age = [data.exog['age'][data.endog == id] for id in party_ID] >>> fig = plt.figure() >>> ax = fig.add_subplot(111) >>> sm.graphics.violinplot(age, ax=ax, labels=labels, ... plot_opts={'cutoff_val':5, 'cutoff_type':'abs', ... 'label_fontsize':'small', ... 'label_rotation':30}) >>> ax.set_xlabel("Party identification of respondent.") >>> ax.set_ylabel("Age") >>> plt.show() .. plot:: plots/graphics_boxplot_violinplot.py Nc6g|]}tj|S)npsize).0arrs ]/var/www/html/test/jupyter/venv/lib/python3.11/site-packages/statsmodels/graphics/boxplots.py zviolinplot..s ) ) )SBGCLL ) ) )rz"No Data to make Violin: Try again!r333333?? violin_width皙?@)notch positionsvert)max ValueErrorr create_mpl_axlistmapr asarrayarangelenmingetzip_single_violinboxplot_set_ticks_labels) dataaxlabelsrside show_boxplot plot_optsmsgfigpos_spanwidthpos_dataposs rrrso` 'YI ) )D ) ) )**a//2oo!"%%GC BJ%% & &DIc$ii((1, vi  26)#4#44H FD268R.111MM.#66;= > >ET9--BB #r3%yAAAA? 4qIA >>>b$ 9=== Jrc Z |dd} fd}tj|}t|| |||} |} || z| z } |dkr | |z| |z} } n.|dkr|| |z} } n |dkr | |z|} } nd} t | ||| | |d d |d d |d d|dd|| fS) bw_factorNcl|dd}|dd}|dd}d}|s |dkr|tj|z}n|}j|z }j|z}tj||dS) zCReturn array with correct range, with which violins can be plotted.cutoffF cutoff_typestd cutoff_valg?gd)r$r r;datasetr#rlinspace) r3r.r9r:r<sx_lowerx_upperkdes r _violin_rangez%_single_violin.._violin_rangesx//mmM599 ]]<55  e##!1!11+//##a'+//##a'{7GS111r) bw_methodrrightleftz<`side` parameter should be one of {'left', 'right', 'both'}. violin_fcz#66c2a5 violin_eck violin_lwr violin_alpha?) facecolor edgecolorlwalpha)r$r r revaluaterr fill_betweenx)r*r4r3r2r,r.r7rDxvalsviolin envelope_l envelope_rr/rCs @rr&r&sb k400I22222"z(##H x9 5 5 5C M(I . .E \\% F V^fjjll *F v~~#)'C-#J "%v|J #)'C-J LooUJ (}}[)DD(}}[#>>! k155$==== ??? &=rc|tj|dz tj|dzg|||d}|d}|s|rddlm}|dt|t|ksd}t|| |} |r || ||r || | dS) z(Set ticks and labels on horizontal axis.rMlabel_fontsizelabel_rotationrsetpNz1Length of `labels` should equal length of `data`.fontsize)rotation) set_xlimr r#r set_xticksr$matplotlib.artistr\r"rset_xticklabels) r*r)r+rr.rYrZr\r/ xticknamess rr(r(sKK ""S("&*;*;c*ABCCCMM)]]#344N]]#344N++****** 6{{c$ii''ECS// !''//  6 Dn 5 5 5 5  6 Dn 5 5 5 5 FrFctj|\}}tttj|}|$t jt|dz}t j|t j |z }t j dt j|dgz| dddz g} t j dt j|dgz| dd dz g} t j dt j|dgz| d d dz g} | d d} t||D]\} }t||| | ||\}}|r|t| |||z}||| d | d d| dd| ddd| dd|  nI|| || z || z| dd | dd| | t!|d} | ddrZ|t j| || z || z| dd| dd| ddrS||t j| | dd | d!d"#t'||||||S)$aH Bean plot of each dataset in a sequence. A bean plot is a combination of a `violinplot` (kernel density estimate of the probability density function per point) with a line-scatter plot of all individual data points. Parameters ---------- data : sequence[array_like] Data arrays, one array per value in `positions`. ax : AxesSubplot If given, this subplot is used to plot in instead of a new figure being created. labels : list[str], optional Tick labels for the horizontal axis. If not given, integers ``1..len(data)`` are used. positions : array_like, optional Position array, used as the horizontal axis of the plot. If not given, spacing of the violins will be equidistant. side : {'both', 'left', 'right'}, optional How to plot the violin. Default is 'both'. The 'left', 'right' options can be used to create asymmetric violin plots. jitter : bool, optional If True, jitter markers within violin instead of plotting regular lines around the center. This can be useful if the data is very dense. plot_opts : dict, optional A dictionary with plotting options. All the options for `violinplot` can be specified, they will simply be passed to `violinplot`. Options specific to `beanplot` are: - 'violin_width' : float. Relative width of violins. Max available space is 1, default is 0.8. - 'bean_color', MPL color. Color of bean plot lines. Default is 'k'. Also used for jitter marker edge color if `jitter` is True. - 'bean_size', scalar. Line length as a fraction of maximum length. Default is 0.5. - 'bean_lw', scalar. Linewidth, default is 0.5. - 'bean_show_mean', bool. If True (default), show mean as a line. - 'bean_show_median', bool. If True (default), show median as a marker. - 'bean_mean_color', MPL color. Color of mean line. Default is 'b'. - 'bean_mean_lw', scalar. Linewidth of mean line, default is 2. - 'bean_mean_size', scalar. Line length as a fraction of maximum length. Default is 0.5. - 'bean_median_color', MPL color. Color of median marker. Default is 'r'. - 'bean_median_marker', MPL marker. Marker type, default is '+'. - 'jitter_marker', MPL marker. Marker type for ``jitter=True``. Default is 'o'. - 'jitter_marker_size', int. Marker size. Default is 4. - 'jitter_fc', MPL color. Jitter marker face color. Default is None. - 'bean_legend_text', str. If given, add a legend with given text. Returns ------- Figure If `ax` is None, the created figure. Otherwise the figure to which `ax` is connected. See Also -------- violinplot : Violin plot, also used internally in `beanplot`. matplotlib.pyplot.boxplot : Standard boxplot. References ---------- P. Kampstra, "Beanplot: A Boxplot Alternative for Visual Comparison of Distributions", J. Stat. Soft., Vol. 28, pp. 1-9, 2008. Examples -------- We use the American National Election Survey 1996 dataset, which has Party Identification of respondents as independent variable and (among other data) age as dependent variable. >>> data = sm.datasets.anes96.load_pandas() >>> party_ID = np.arange(7) >>> labels = ["Strong Democrat", "Weak Democrat", "Independent-Democrat", ... "Independent-Indpendent", "Independent-Republican", ... "Weak Republican", "Strong Republican"] Group age by party ID, and create a violin plot with it: >>> plt.rcParams['figure.subplot.bottom'] = 0.23 # keep labels visible >>> age = [data.exog['age'][data.endog == id] for id in party_ID] >>> fig = plt.figure() >>> ax = fig.add_subplot(111) >>> sm.graphics.beanplot(age, ax=ax, labels=labels, ... plot_opts={'cutoff_val':5, 'cutoff_type':'abs', ... 'label_fontsize':'small', ... 'label_rotation':30}) >>> ax.set_xlabel("Party identification of respondent.") >>> ax.set_ylabel("Age") >>> plt.show() .. plot:: plots/graphics_boxplot_beanplot.py Nrrrrrr bean_sizerMbean_mean_sizebean_legend_textr6 jitter_markerojitter_marker_size bean_colorrJ jitter_fcnone)lsmarkermsmecmewmfclabelbean_lw)rPcolorrvbean_show_meanT bean_mean_lwbean_mean_colorb)rPrxbean_show_medianbean_median_marker+bean_median_colorr)rqrx)rrrrr r r!r"rr#r$r%r&_jitter_envelopeplothlines _show_legendmeanmedianr()r)r*r+rr,jitterr.r0r1r bean_widthbean_mean_width legend_txtr3r4rTrU jitter_coords rrrsH!"%%GC BJ%% & &DIc$ii((1, vi  26)#4#44H64"&(B"8"88MM.#66;=>>L"~ 6 66MM+s33b8:;;JfdRVXrN%;%;;MM"2C882=?@@O1488JT9--#C#C #&r3,iXX v  (!1(E64!P!PPL GGL(r$==#>> }}%91==! lC88y}}[&AA$  & & & & IIhj 0# 2B"y#66%MM,<<&  ( ( (  !    J ==)4 0 0 C IIbgh'')>o@U"~r::%MM*;SAA  C C C ==+T 2 2 C GGC8,,$==)=sCC#--(;SAA  C C C b$ 9=== Jrc|dkrd\}}n*|dkrd\}}n|dkrd\}}ntd|ztj|||}|tj|||jz}|S) z&Determine envelope for jitter markers.r)rrF)rrrG)rrz`side` input incorrect: %s)lowhighr )rr interprandomuniformr )r3rTrUr,rrjitter_envelopers rrrs v~~ TT  TT  TT5<===i%88O"RY%6%63Trs $$$$$$  $AG,0JJJJZ///d   644fR\\\\~$     r