Skip to content

Commit

Permalink
Merge pull request #13 from kamilm/unstable
Browse files Browse the repository at this point in the history 
Boxplot addition and histogram fix
  • Loading branch information
@kamilm
kamilm committed Dec 27, 2014
2 parents 6387bbd + f96251b commit bfeaf35
Show file tree
Hide file tree
Showing 2 changed files with 179 additions and 66 deletions.
222 changes: 158 additions & 64 deletions kaplot/__init__.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -3,6 +3,10 @@
generation potention of matplotlib with an easier to use, object oriented, interface. The interface
is simple enough to quickly prototype plots, or fine tune for publication quality results.
NOTES
=====
- plot_type = boxplot : kwargs passed to any iteration of add_plotdata() will be used for all instances of add_plotdata
TODO
====
- sanitize linestyle/ls = '-',... vs 'solid','dashed','dashdot','dotted'
Expand All @@ -28,7 +32,7 @@
from numpy import linspace

__author__ = 'Kamil'
__version__ = '1.0.0~beta5'
__version__ = '1.0.0~beta6'
__name__ = 'kaplot'

@decorator
Expand Down Expand Up @@ -183,9 +187,11 @@ def set_plot_type(self,ptype,**kwargs):
name - layer name if not main
"""
k = self._LAYER_OBJECTS[kwargs['ind']]
if ptype.lower() in ['line', 'bar','hist','histogram']:
if ptype.lower() in ['line', 'bar','hist','histogram', 'box', 'boxplot']:
if ptype.lower() in ['hist', 'histogram']:
ptype = 'hist'
if ptype.lower() in ['box', 'boxplot']:
ptype = 'boxplot'
k.set_plot_type(ptype)
else:
print "KAPLOT Error. Not a valid plot"
Expand Down Expand Up @@ -766,6 +772,27 @@ def add_plotdata(self,x,y,**kwargs):
hatch -
ls - line style
lw - line width
** boxplot chart kwargs **
http://matplotlib.org/api/pyplot_api.html#matplotlib.pyplot.boxplot
vert - True/False : plot verical or horizontal
whis - 1.5 : size of the whiskers, $whis$ * Quartile
loc - axes location : maps to positions (array)
width - width of box : maps to widths (array)
meanline - True/False : show a mean as a line
showmean - True/False : show the mean according to meanprops (maps to showmeans)
showcap - True/False : show the caps at the end of whiskers (maps to showcaps)
showbox - True/False : show the box portion of boxplot
showfliers - True/False : show the outliers
boxprop - dictionary : properties for the box
label - the label : maps to labels (array)
flierprops - dictionary : properties for fliers
medianprops - dictionary : properties for median line
meanprops - dictionary : properties for the mean line
capprops - dictionary : properties for the caps
whiskerprops- dictionary : properties for the whisker lines
manage_xticks- True/False : whether or not 'label' is given to the tick
# TODO : test manage ticks
"""
k = self._LAYER_OBJECTS[kwargs['ind']]
kwargs['x'] = x
Expand Down Expand Up @@ -931,69 +958,135 @@ def color_marker_fill_index(cnt,clist,mlist,flist):
if 'max' in npd.keys():
npd['range'][1] = npd.pop('max')
k.DATA_LIST[i] = npd
# generate color,marker,fill list for the plot
inc_cnt = 0
for pd in k.DATA_LIST:
if pd['increment']:
inc_cnt += 1
cnt = 0
for pd in k.DATA_LIST:
# line plots
if k.SETTINGS['plot_type'] == 'line':
if k.SETTINGS['uniq_cols']:
cols = unique_colors(inc_cnt+1,k.SETTINGS['color_map'])
col , mar , fill = cols[cnt] , None , None
else:
cind , mind , find = color_marker_fill_index(cnt,self._COLOR_LIST,self._MARKER_LIST,self._MARKER_FILL_LIST)
col , mar , fill = self._COLOR_LIST[cind] , self._MARKER_LIST[mind] , self._MARKER_FILL_LIST[find]
if pd['increment']:
cnt += 1
if 'color' not in pd:
pd['color'] = col
if 'marker' not in pd:
pd['marker'] = mar
if 'mfc' not in pd:
pd['mfc'] = fill
# spline portion
sp_key = ['color','lw','ls']
if pd['spline']:
x_spline = linspace(pd['x'][0],pd['x'][-1],pd['sp_points'])
y_spline = UnivariateSpline(pd['x'],pd['y'],k=pd['sp_order'],s=pd['sp_smooth'])(x_spline)
sp_dict = {}
for sp in sp_key:
if sp in pd:
sp_dict[sp] = pd[sp]
pd['lw'] = 0
pd['ls'] = ''
mpobj.errorbar(x=x_spline,y=y_spline,**sp_dict)
pd.pop('spline')
pd.pop('sp_smooth')
pd.pop('sp_order')
pd.pop('sp_points')
pd.pop('increment')
mpobj.errorbar(**pd)
# bar plots
elif k.SETTINGS['plot_type'] in ['bar', 'hist']:
if k.SETTINGS['uniq_cols']:
cols = unique_colors(inc_cnt+1,k.SETTINGS['color_map'])
col , hat , fill = cols[cnt] , None , None
else:
cind , hind , find = color_marker_fill_index(cnt,self._COLOR_LIST,self._HATCH_LIST,self._HATCH_FILL_LIST)
col , hat , fill = self._COLOR_LIST[cind] , self._HATCH_LIST[hind] , self._HATCH_FILL_LIST[find]
elif k.SETTINGS['plot_type'] == 'boxplot':
npd = update_default_kwargs(self._BOXPLOT_DEFAULTS,pd)
npd['x'] = pd['y']
k.DATA_LIST[i] = npd
if k.SETTINGS['plot_type'] in ['line','bar']:
# generate color,marker,fill list for the plot
inc_cnt = 0
for pd in k.DATA_LIST:
if pd['increment']:
cnt += 1
# do not overwrite user specified values
if 'color' not in pd:
pd['color'] = col
if 'hatch' not in pd:
pd['hatch'] = hat
if 'fill' not in pd:
pd['fill'] = fill
pd.pop('increment')
if k.SETTINGS['plot_type'] == 'bar':
inc_cnt += 1
cnt = 0
for pd in k.DATA_LIST:
# line plots
if k.SETTINGS['plot_type'] == 'line':
if k.SETTINGS['uniq_cols']:
cols = unique_colors(inc_cnt+1,k.SETTINGS['color_map'])
col , mar , fill = cols[cnt] , None , None
else:
cind , mind , find = color_marker_fill_index(cnt,self._COLOR_LIST,self._MARKER_LIST,self._MARKER_FILL_LIST)
col , mar , fill = self._COLOR_LIST[cind] , self._MARKER_LIST[mind] , self._MARKER_FILL_LIST[find]
if pd['increment']:
cnt += 1
if 'color' not in pd:
pd['color'] = col
if 'marker' not in pd:
pd['marker'] = mar
if 'mfc' not in pd:
pd['mfc'] = fill
# spline portion
sp_key = ['color','lw','ls']
if pd['spline']:
x_spline = linspace(pd['x'][0],pd['x'][-1],pd['sp_points'])
y_spline = UnivariateSpline(pd['x'],pd['y'],k=pd['sp_order'],s=pd['sp_smooth'])(x_spline)
sp_dict = {}
for sp in sp_key:
if sp in pd:
sp_dict[sp] = pd[sp]
pd['lw'] = 0
pd['ls'] = ''
mpobj.errorbar(x=x_spline,y=y_spline,**sp_dict)
pd.pop('spline')
pd.pop('sp_smooth')
pd.pop('sp_order')
pd.pop('sp_points')
pd.pop('increment')
mpobj.errorbar(**pd)
# bar plots
elif k.SETTINGS['plot_type'] in ['bar']:
if k.SETTINGS['uniq_cols']:
cols = unique_colors(inc_cnt+1,k.SETTINGS['color_map'])
col , hat , fill = cols[cnt] , None , None
else:
cind , hind , find = color_marker_fill_index(cnt,self._COLOR_LIST,self._HATCH_LIST,self._HATCH_FILL_LIST)
col , hat , fill = self._COLOR_LIST[cind] , self._HATCH_LIST[hind] , self._HATCH_FILL_LIST[find]
if pd['increment']:
cnt += 1
# do not overwrite user specified values
if 'color' not in pd:
pd['color'] = col
if 'hatch' not in pd:
pd['hatch'] = hat
if 'fill' not in pd:
pd['fill'] = fill
pd.pop('increment')
mpobj.bar(**pd)
else:
mpobj.hist(**pd)
elif k.SETTINGS['plot_type'] in ['hist','boxplot']:
if k.SETTINGS['plot_type'] == 'hist':
x_list = []
labels = []
colors = []
histargs = {}
for i,pd in enumerate(k.DATA_LIST):
if k.SETTINGS['uniq_cols']:
cols = unique_colors(inc_cnt+1,k.SETTINGS['color_map'])
col = cols[cnt]
else:
cind , hind , find = color_marker_fill_index(cnt,self._COLOR_LIST,self._HATCH_LIST,self._HATCH_FILL_LIST)
col = self._COLOR_LIST[cind]
if pd['increment']:
cnt += 1
pd.pop('increment')
# do not overwrite user specified values
if 'color' not in pd:
colors.append(col)
else:
colors.append(pd['color'])
pd.pop('color')
# data addition
x_list.append(pd['x'])
pd.pop('x')
# data labels
if 'label' in pd:
labels.append(pd['label'])
pd.pop('label')
else:
labels.append('')
# build large plot args
for key,val in pd.iteritems():
histargs[key] = val
mpobj.hist(x=x_list,label=labels,color=colors,**histargs)
elif k.SETTINGS['plot_type'] == 'boxplot':
x_list = []
labels = []
positions = []
bpargs = {}
for i,pd in enumerate(k.DATA_LIST):
# add data to plot
x_list.append(pd['x'])
# pop off the values that are not required anymore.
pd.pop('x')
pd.pop('increment')
# add labels to the data sets
if 'label' in pd:
labels.append(pd['label'])
pd.pop('label')
else:
labels.append(None)
# customize the positions
if 'loc' in pd:
positions.append(pd['loc'])
pd.pop('loc')
else:
positions.append(i+1)
# update bpargs with user passed variabls and preform rename if required
for key,val in pd.iteritems():
if key in ['width','showmean','showcap']:
key = key+'s'
bpargs[key] = val
mpobj.boxplot(x=x_list,labels=labels,positions=positions,**bpargs)
# AXES LABELS, TICKS, FORMATTING, and PARAMETERS
if k.SETTINGS['xlabel'] is not None:
mpobj.set_xlabel(k.SETTINGS['xlabel'],**k.SETTINGS['xlab_prop'])
Expand Down Expand Up @@ -1111,7 +1204,8 @@ def color_marker_fill_index(cnt,clist,mlist,flist):
k.SETTINGS['leg_props'].pop('bool')
l = mpobj.legend(prop=k.SETTINGS['leg_fprop'],**k.SETTINGS['leg_props'])
# update the legend title also
plt.setp(l.get_title(),**k.SETTINGS['leg_fprop'])
if k.SETTINGS['leg_props']['title'] is not None:
plt.setp(l.get_title(),**k.SETTINGS['leg_fprop'])
# make copy of the entire object
self._LAYER_PLT_OBJECT.append(mpobj)
return
Expand Down
23 changes: 21 additions & 2 deletions kaplot/defaults.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -181,8 +181,8 @@
'log' : 'Auto' , \
'alpha' : 'Auto'},

'_HIST_DEFAULTS' : { 'bins' : 'Auto' , \
'increment' : True , \
'_HIST_DEFAULTS' : { 'increment' : True , \
'bins' : 'Auto' , \
'min' : 'Auto' , \
'max' : 'Auto' , \
'normed' : 'Auto' , \
Expand All @@ -202,6 +202,25 @@
'ls' : 'Auto' , \
'lw' : 'Auto'},

'_BOXPLOT_DEFAULTS' : { 'increment' : True , \
'label' : 'Auto' , \
'loc' : 'Auto' , \
'whis' : 'Auto' , \
'vert' : 'Auto' , \
'width' : 'Auto' , \
'meanline' : 'Auto' , \
'showmean' : 'Auto' , \
'showcap' : 'Auto' , \
'showbox' : 'Auto' , \
'showfliers' : 'Auto' , \
'boxprops' : 'Auto' , \
'flierprops' : 'Auto' , \
'medianprops' : 'Auto' , \
'meanprops' : 'Auto' , \
'capprops' : 'Auto' , \
'whiskerprops' : 'Auto' , \
'manage_xticks' : 'Auto'},

'_LEGEND_FONTPROPS' : { 'family' : 'sans-serif' , \
'weight' : 'normal' , \
'size' : 'medium'},
Expand Down

0 comments on commit bfeaf35

Please sign in to comment.