def test_irregular_subplots():
df = cf.datagen.bubble(10, 50, mode='stocks')
figs = cf.figures(df, [
dict(kind='histogram', keys='x', color='blue'),
dict(kind='scatter', mode='markers', x='x', y='y', size=5),
dict(
kind='scatter', mode='markers', x='x', y='y', size=5, color='teal')
],
asList=True)
figs.append(
cf.datagen.lines(1).figure(bestfit=False,
colors=['blue'],
bestfit_colors=['pink']))
base_layout = cf.tools.get_base_layout(figs)
sp = cf.subplots(
figs,
shape=(3, 2),
base_layout=base_layout,
vertical_spacing=.15,
horizontal_spacing=.03,
specs=[[{
'rowspan': 2
}, {}], [None, {}], [{
'colspan': 2
}, None]],
subplot_titles=['Histogram', 'Scatter 1', 'Scatter 2', 'Bestfit Line'])
sp['layout'].update(showlegend=False)
return sp
def test_irregular_subplots():
df = cf.datagen.bubble(10, 50, mode='stocks')
figs = cf.figures(df, [
dict(kind='histogram', keys='x', color='blue'),
dict(kind='scatter', mode='markers', x='x', y='y', size=5),
dict(kind='scatter', mode='markers', x='x', y='y',
size=5, color='teal')],asList=True)
figs.append(cf.datagen.lines(1).figure(bestfit=False, colors=['blue'],
bestfit_colors=['pink']))
base_layout = cf.tools.get_base_layout(figs)
sp = cf.subplots(figs, shape=(3, 2), base_layout=base_layout,
vertical_spacing=.15, horizontal_spacing=.03,
specs=[[{'rowspan': 2}, {}], [None, {}],
[{'colspan': 2}, None]],
subplot_titles=['Histogram', 'Scatter 1',
'Scatter 2', 'Bestfit Line'])
sp['layout'].update(showlegend=False)
return sp
def test_irregular_subplots():
df = cf.datagen.bubble(10, 50, mode="stocks")
figs = cf.figures(
df,
[
dict(kind="histogram", keys="x", color="blue"),
dict(kind="scatter", mode="markers", x="x", y="y", size=5),
dict(kind="scatter", mode="markers", x="x", y="y", size=5, color="teal"),
],
asList=True,
)
figs.append(cf.datagen.lines(1).figure(bestfit=False, colors=["blue"], bestfit_colors=["pink"]))
base_layout = cf.tools.get_base_layout(figs)
sp = cf.subplots(
figs,
shape=(3, 2),
base_layout=base_layout,
vertical_spacing=0.15,
horizontal_spacing=0.03,
specs=[[{"rowspan": 2}, {}], [None, {}], [{"colspan": 2}, None]],
subplot_titles=["Histogram", "Scatter 1", "Scatter 2", "Bestfit Line"],
)
sp["layout"].update(showlegend=False)
return sp
def plot_chart(self, data_frame, style, chart_type):
mode = 'line'
if style is None: style = Style()
marker_size = 1
x = ''
y = ''
z = ''
fig = None
scale = 1
try:
if (style.plotly_plot_mode == 'offline_html'):
scale = 2 / 3
except:
pass
# check other plots implemented by Cufflinks
if fig is None:
cm = ColorMaster()
# create figure
data_frame_list = self.split_data_frame_to_list(data_frame, style)
fig_list = []
cols = []
for data_frame in data_frame_list:
cols.append(data_frame.columns)
cols = list(np.array(cols).flat)
# get all the correct colors (and construct gradients if necessary eg. from 'Blues')
# need to change to strings for cufflinks
color_list = cm.create_color_list(style, [], cols=cols)
color_spec = []
# if no colors are specified then just use our default color set from chart constants
if color_list == [None] * len(color_list):
color_spec = [None] * len(color_list)
for i in range(0, len(color_list)):
# get the color
if color_spec[i] is None:
color_spec[i] = self.get_color_list(i)
try:
color_spec[i] = matplotlib.colors.rgb2hex(
color_spec[i])
except:
pass
else:
# otherwise assume all the colors are rgba
for color in color_list:
color = 'rgba' + str(color)
color_spec.append(color)
start = 0
for i in range(0, len(data_frame_list)):
data_frame = data_frame_list[i]
if isinstance(chart_type, list):
chart_type_ord = chart_type[i]
else:
chart_type_ord = chart_type
end = start + len(data_frame.columns)
color_spec1 = color_spec[start:start + end]
start = end
if chart_type_ord == 'surface':
fig = data_frame.iplot(
kind=chart_type,
title=style.title,
xTitle=style.x_title,
yTitle=style.y_title,
x=x,
y=y,
z=z,
mode=mode,
size=marker_size,
theme=style.plotly_theme,
bestfit=style.line_of_best_fit,
legend=style.display_legend,
colorscale=style.color,
dimensions=(style.width * style.scale_factor * scale,
style.height * style.scale_factor * scale),
asFigure=True)
elif chart_type_ord == 'line':
chart_type_ord = 'scatter'
elif chart_type_ord == 'scatter':
mode = 'markers'
marker_size = 5
elif chart_type_ord == 'bubble':
x = data_frame.columns[0]
y = data_frame.columns[1]
z = data_frame.columns[2]
# special case for choropleth which has yet to be implemented in Cufflinks
# will likely remove this in the future
elif chart_type_ord == 'choropleth':
for col in data_frame.columns:
try:
data_frame[col] = data_frame[col].astype(str)
except:
pass
if style.color != []:
color = style.color
else:
color = [[0.0, 'rgb(242,240,247)'], [0.2, 'rgb(218,218,235)'], [0.4, 'rgb(188,189,220)'], \
[0.6, 'rgb(158,154,200)'], [0.8, 'rgb(117,107,177)'], [1.0, 'rgb(84,39,143)']]
text = ''
if 'text' in data_frame.columns:
text = data_frame['Text']
data = [
dict(type='choropleth',
colorscale=color,
autocolorscale=False,
locations=data_frame['Code'],
z=data_frame[
style.plotly_choropleth_field].astype(float),
locationmode=style.plotly_location_mode,
text=text,
marker=dict(
line=dict(color='rgb(255,255,255)', width=1)),
colorbar=dict(title=style.units))
]
layout = dict(
title=style.title,
geo=dict(
scope=style.plotly_scope,
projection=dict(type=style.plotly_projection),
showlakes=True,
lakecolor='rgb(255, 255, 255)',
),
)
fig = dict(data=data, layout=layout)
if chart_type_ord not in ['surface', 'choropleth']:
fig = data_frame.iplot(
kind=chart_type_ord,
title=style.title,
xTitle=style.x_title,
yTitle=style.y_title,
x=x,
y=y,
z=z,
subplots=False,
mode=mode,
size=marker_size,
theme=style.plotly_theme,
bestfit=style.line_of_best_fit,
legend=style.display_legend,
color=color_spec1,
dimensions=(style.width * style.scale_factor * scale,
style.height * style.scale_factor * scale),
asFigure=True)
fig.update(dict(layout=dict(legend=dict(x=0.05, y=1))))
fig.update(dict(layout=dict(paper_bgcolor='rgba(0,0,0,0)')))
fig.update(dict(layout=dict(plot_bgcolor='rgba(0,0,0,0)')))
fig_list.append(fig)
if len(fig_list) > 1:
import cufflinks
fig = cufflinks.subplots(fig_list)
else:
fig = fig_list[0]
self.publish_plot(fig, style)
#!/usr/bin/python3
import pandas as pd
import numpy as np
from optparse import OptionParser
import pathos.multiprocessing as mp
import cufflinks
import plotly as py
from tqdm import tqdm
import time
pd.set_option('expand_frame_repr', False)
df=pd.read_csv('del.csv',index_col=[0])
pvt=df.pivot_table(values='mse',columns='method', index=['quant_size','number_of_bins','std_threshold'],aggfunc=[lambda x: x.count()/len(x),np.mean]).rename(columns={'<lambda>':'pass','mean':'mse'})
reshape=x=pvt.stack().reset_index(drop=False).pivot_table(index=['pass'],values=['mse'],columns=['method','number_of_bins'])
if 0:
reshape.columns=reshape.columns.swaplevel(0,2)
py.offline.plot(reshape[(23,'mse_from_modulo_method')].dropna().sort_index().iplot(asFigure=True))
else:
py.offline.plot(reshape.mse.mse_from_modulo_method.dropna().sort_index().iplot(asFigure=True))
exit()
columns=[i[:-1] for i in reshape.columns]
f2=cufflinks.subplots([reshape[i].figure() for i in columns],subplot_titles=[str(i) for i in columns])
print('hi')
py.offline.plot(f2)
df.iplot(subplots=True, subplot_titles=True, legend=False)
df = cf.datagen.bubble(10, 50, mode='stocks')
figs = cf.figures(df, [
dict(kind='histogram', keys='x', color='blue'),
dict(kind='scatter', mode='markers', x='x', y='y', size=5),
dict(kind='scatter', mode='markers', x='x', y='y', size=5, color='teal')
],
asList=True)
figs.append(
cf.datagen.lines(1).figure(bestfit=True,
colors=['blue'],
bestfit_colors=['pink']))
base_layout = cf.tools.get_base_layout(figs)
sp = cf.subplots(
figs,
shape=(3, 2),
base_layout=base_layout,
vertical_spacing=.15,
horizontal_spacing=.03,
specs=[[{
'rowspan': 2
}, {}], [None, {}], [{
'colspan': 2
}, None]],
subplot_titles=['Histogram', 'Scatter 1', 'Scatter 2', 'Bestfit Line'])
sp['layout'].update(showlegend=False)
cf.iplot(sp)
help(df.iplot)
title='up视频质量走势参考',
xTitle='时间',
yTitle='比率')
return df3.figure(
kind='lines',
orientation='v',
)
def month_video_play(self):
df1 = self.df_data[[
'created', 'stat#favorite', 'stat#coin', 'stat#like'
]]
df2 = df1.groupby([df1.created.dt.year,
df1.created.dt.month]).agg('sum')
# df2.iplot(kind='bar', barmode='stack', orientation='v')
df2.iplot(kind='histogram', bins=10)
return df2
if __name__ == '__main__':
upid = 176037767
analysis_video = AnalysisVideo(upid)
figs = cf.figures(analysis_video.month_video_play(),
[dict(kind='bar', barmode='stack')],
asList=True)
figs.append(analysis_video.month_video_play_rate())
layout = cf.tools.get_base_layout(figs)
sp = cf.subplots(figs, shape=(1, 2), base_layout=layout)
sp['layout'].update(showlegend=False)
cf.iplot(sp)
# In[9]:
train[['Sex', 'Survived']].groupby('Sex').mean().iplot(kind='bar', colors='blue', title='Survival Rate based on Gender')
# Around 74% females in the sample survived compared to 18% males.
# Our 1st assumption seems to hold true.
# In[10]:
survived = train[train['Survived'] == 1]['Age']
not_survived = train[train['Survived'] == 0]['Age']
cf.subplots([survived.figure(kind='hist', colors='blue'), not_survived.figure(kind='hist', colors='blue')],
shared_yaxes=True, subplot_titles=['Survived', 'Not Survived']).iplot()
# The histogram above shows us that survival rate of children is more compared to adults. <br>
# Few more observations:
# * Infant (Age < 5) survival rate is more <br>
# * Most of the passengers aged between 15 to 35 did not survived <br>
# * Very few of the oldest passengers (Age=80) survived
# In[11]:
train[['Pclass', 'Survived']].groupby('Pclass').mean().iplot(kind='bar', colors='blue', title='Survival Rate based on Passenger Class')
# Class 1 passengers clearly had the highest survival rate followed by Class 2 and 3
def show(self):
self.figure = cf.Figure(
cf.subplots(self.figure_list, base_layout=self.figure_list[1].to_dict()['layout'], shared_xaxes=True, shape=(len(self.figure_list), 1)))
self.figure.show()