Related
I have a plotly plot which looks like this:
The Code I am using is below:
fig = make_subplots(specs=[[{"secondary_y": True}]])
fig.add_trace(go.Scatter( x = pf['Timestamp'], y = pf['Price_A'], name ='<b>A</b>',
mode = 'lines+markers',
marker_color = 'rgba(255, 0, 0, 0.8)',
line = dict(width = 3 ), yaxis = "y1"),
secondary_y=False,)
fig.add_trace(go.Scatter( x = df['Timestamp'], y = df['Price_B'], name='<b>B</b>',
mode = 'lines+markers',
marker_color = 'rgba(0, 196, 128, 0.8)',
line = dict(width = 3 ), yaxis = "y1") ,
secondary_y=False,)
for i in pf2['Timestamp']:
fig.add_vline(x=i, line_width=3, line_dash="dash", line_color="purple",
name='Event')
fig.update_layout( title="<b>Change over Time</b>", font=dict( family="Courier New,
monospace", size=16, color="RebeccaPurple"),
legend=dict(
yanchor="top",
y=0.99,
xanchor="left",
x=0.01
))
How can I add the entry in the legend for the event that is denoted by the vertical lines?
When you use add_vline, you are adding an annotation which will not have a corresponding legend entry.
You'll need to instead use go.Scatter to plot the vertical lines, passing the minimum and maximum values in your data (plus or minus some padding) to the y parameter. Then you can set this same y-range for your plot. This will give you the appearance of vertical lines while still showing the full range of your data.
Update: you can use a legend group so that the vertical lines appear as a single entry in the legend
For example:
from pkg_resources import yield_lines
import plotly.express as px
import plotly.graph_objects as go
fig = go.Figure()
df = px.data.stocks()
for col in ['GOOG','AMZN']:
fig.add_trace(go.Scatter(
x=df['date'],
y=df[col]
))
vlines = ["2018-07-01","2019-04-01","2019-07-01"]
min_y,max_y = df[['GOOG','AMZN']].min().min(), df[['GOOG','AMZN']].max().max()
padding = 0.05*(max_y-min_y)
for i,x in enumerate(vlines):
fig.add_trace(go.Scatter(
x=[x]*2,
y=[min_y-padding, max_y+padding],
mode='lines',
line=dict(color='purple', dash="dash"),
name="vertical lines",
legendgroup="vertical lines",
showlegend=True if i == 0 else False
))
fig.update_yaxes(range=[min_y-padding, max_y+padding])
fig.show()
I am having difficulties with setting an equal space between pie charts of different sizes. The 5 are correctly arranged in one row, but the distance between the contours of neighboring pies aren't equal. I tried many abbreviations of the following code, all of them not making a big difference in the output (see image):
#code:
import matplotlib.pyplot as plt
import pandas as pd
labels = 'Verkehr', 'Maschinen und Motoren', 'Feuerungen', 'Industrie / Gewerbe', 'Land- und Forstwirtschaft'
sizesax1 = [108295, 10107, 7220, 11551, 7220]
sizesax2 = [77882, 6676, 6676, 13351, 6676]
sizesax3 = [55652, 4417, 6184, 15900, 6184]
sizesax4 = [36327, 2642, 4632, 16512, 5944]
sizesax5 = [18781, 1409, 3287, 1878, 4695]
fig, (ax1, ax2, ax3, ax4, ax5) = plt.subplots(1, 5, figsize =(20,4))
ax1.pie(sizesax1, startangle=0, colors = ('red', 'darkblue', 'orange', 'yellow', 'green'), radius=1*4)
ax2.pie(sizesax2, startangle=0, colors = ('red', 'darkblue', 'orange', 'yellow', 'green'), radius=.77*4)
ax3.pie(sizesax3, startangle=0, colors = ('red', 'darkblue', 'orange', 'yellow', 'green'), radius=.61*4)
ax4.pie(sizesax4, startangle=0, colors = ('red', 'darkblue', 'orange', 'yellow', 'green'), radius=.46*4)
ax5.pie(sizesax5, startangle=0, colors = ('red', 'darkblue', 'orange', 'yellow', 'green'), radius=.33*4)
some additions i tried:
fig.subplots_adjust(left=None, bottom=None, right=None, top=None, wspace=1, hspace=None)
or
fig.tight_layout()
#giving me this error message:
/srv/conda/envs/notebook/lib/python3.7/site-packages/ipykernel_launcher.py:17: UserWarning:
Tight layout not applied. The bottom and top margins cannot be made large enough to
accommodate all axes decorations.
and some others.
Big thank you already for reading this! I am a complete beginner in python and just managed to come as far as you see in this image:
enter image description here
It is not clear what it is required. I'll assume it is the following image:
Fundamentally, the problem is that the pie needs a square aspect ratio, which is not provided by a row of subplots.
The simplest solution, is to create only one plot and plot there multiple pies with different centres. Something like:
import matplotlib.pyplot as plt
sizes = [ [108295, 10107, 7220, 11551, 7220],
[77882, 6676, 6676, 13351, 6676],
[55652, 4417, 6184, 15900, 6184],
[36327, 2642, 4632, 16512, 5944],
[18781, 1409, 3287, 1878, 4695]]
colors = ('red', 'darkblue', 'orange', 'yellow', 'green')
R = 4
radius = [R*i for i in [1.0, 0.77, 0.61, 0.46, 0.33] ]
wid = sum(radius)*2
hei = R*2
fig, ax = plt.subplots(figsize =(wid,hei))
fig.subplots_adjust(left = 0, right = 1, bottom = 0, top = 1)
y = R
x = 0
for i in range(5):
x += radius[i]
ax.pie(sizes[i], startangle = 0, colors = colors,
radius = radius[i], center = (x,y) )
x += radius[i]
ax.set(xlim =(0,x), ylim=(0,R*2))
plt.savefig("aaa.png")
Notice that my figure aspect ratio is not the (20,4) of the question, which does not hold for the way I interpreted the intended result.
But it might be the case that there is the need of having these in different axes. If so, the idea is:
Use gridspec to create a single row with 5 columns and provide the ratios so that they correspond to the required radius.
Plot the larger pie in the left slot.
In all remaining slots, use a subgrid, dividing into a column of three (sub-)slots.
Set the height ratios so that the middle one ends up with an aspect ratio of a square.
Plot the pies in the middle slots.
Here we go:
import matplotlib.pyplot as plt
sizes = [ [108295, 10107, 7220, 11551, 7220],
[77882, 6676, 6676, 13351, 6676],
[55652, 4417, 6184, 15900, 6184],
[36327, 2642, 4632, 16512, 5944],
[18781, 1409, 3287, 1878, 4695]]
colors = ('red', 'darkblue', 'orange', 'yellow', 'green')
R = 4
radius = [R*i for i in [1.0, 0.77, 0.61, 0.46, 0.33] ]
wid = sum(radius)*2
hei = R*2
ratios = [i/radius[0] for i in radius] # for gridspec
fig = plt.figure(figsize =(wid,hei))
gs = fig.add_gridspec(1, 5,
width_ratios = ratios,
wspace=0, left = 0, right = 1, bottom = 0, top = 1)
ax = fig.add_subplot(gs[0,0])
ax.pie(sizes[0], startangle = 0, colors = colors, radius = 1 )
ax.set(xlim=(-1,1) ,ylim=(-1,1))
for i in range(1,5):
mid = ratios[i]/sum(ratios)*wid
inrat = [(hei-mid)/2, mid, (hei-mid)/2]
ings = gs[0,i].subgridspec(3, 1, hspace=0,
height_ratios = inrat)
ax = fig.add_subplot(ings[1,0])
ax.pie(sizes[i], startangle = 0, colors = colors, radius = 1 )
ax.set(xlim=(-1,1), ylim=(-1,1))
plt.savefig("aaa.png")
I am trying to achieve what is done here: https://www.quantalys.com/Fonds/120955 with javascript in python plotly. I want to add the hover vertical line and the red annotation on the x axis. I have done some searching on goolgle but I couldn't find the the answer I'm looking for. My current chart looks like this:
trace1 = go.Scatter(
x = df1.x,
y = df1.y,
name = "M&G OPTIMAL INCOME FD EUR AH ACC",
hoverinfo= 'name',
opacity=0.7,
mode = 'lines',
line = dict(
color = ('rgb(2, 12, 245)'),
width = 1,
),
)
trace2 = go.Scatter(
x = df2.x,
y = df2.y,
opacity=0.7,
name = "Alloc Flexible Prudent Monde",
hoverinfo= 'name',
mode = 'lines',
line = dict(
color = ('rgb(67, 45, 24)'),
width = 1,
)
)
trace3 = go.Scatter(
x = df3.x,
y = df3.y,
name = "25% MSCI World + 75% ML Global",
hoverinfo= 'name',
mode = 'lines',
opacity=0.7,
line = dict(
color = ('rgb(205, 12, 24)'),
width = 1,
)
)
layout = go.Layout(
xaxis=dict(
showline=True,
showgrid=True,
showticklabels=True,
linecolor='rgb(204, 204, 204)',
linewidth=2,
mirror=True,
),
yaxis=dict(
showline=True,
showgrid=True,
showticklabels=True,
linecolor='rgb(204, 204, 204)',
linewidth=2,
mirror=True,
),
showlegend=True,
)
data= [trace1, trace2,trace3]
fig = dict(data=data, layout=layout)
iplot(fig, filename='line-mode')
Add this to your layout definition.
showlegend = True,
hovermode = 'x'
Add this to your xaxis definition.
showspikes = True,
spikemode = 'across',
spikesnap = 'cursor',
showline=True,
showgrid=True,
...
And add this to your layout definition:
spikedistance = -1,
xaxis=dict(...
Please refer to this post and the documentation by plotly. :)
EDIT
You ask for the x-axis lable. Please use
spikemode = 'across+toaxis'
Additionally I would suggest to use
spikedash = 'solid'
because it is better fitting your example.
I just want to create a menu on the plot where I'm able to change the z-value in data only. I tried looking at other examples on here: https://plot.ly/python/dropdowns/#restyle-dropdown but it was hard since the examples were not exactly similar to my plot.
import plotly
import plotly.plotly as py
import plotly.graph_objs as go
import pandas as pd
df = pd.read_csv('https://raw.githubusercontent.com/plotly/datasets/master/2014_world_gdp_with_codes.csv')
data = [go.Choropleth(
locations = df['CODE'],
z = df['GDP (BILLIONS)'],
text = df['COUNTRY'],
colorscale = [
[0, "rgb(5, 10, 172)"],
[0.35, "rgb(40, 60, 190)"],
[0.5, "rgb(70, 100, 245)"],
[0.6, "rgb(90, 120, 245)"],
[0.7, "rgb(106, 137, 247)"],
[1, "rgb(220, 220, 220)"]
],
autocolorscale = False,
reversescale = True,
marker = go.choropleth.Marker(
line = go.choropleth.marker.Line(
color = 'rgb(180,180,180)',
width = 0.5
)),
colorbar = go.choropleth.ColorBar(
tickprefix = '$',
title = 'GDP<br>Billions US$'),
)]
layout = go.Layout(
title = go.layout.Title(
text = '2014 Global GDP'
),
geo = go.layout.Geo(
showframe = False,
showcoastlines = False,
projection = go.layout.geo.Projection(
type = 'equirectangular'
)
),
annotations = [go.layout.Annotation(
x = 0.55,
y = 0.1,
xref = 'paper',
yref = 'paper',
text = 'Source: <a href="https://www.cia.gov/library/publications/the-world-factbook/fields/2195.html">\
CIA World Factbook</a>',
showarrow = False
)]
)
fig = go.Figure(data = data, layout = layout)
py.iplot(fig, filename = 'd3-world-map')
It's been a while since this was asked, but I figured it was still worth answering. I can't speak to how this might have changed since it was asked in 2019, but this works today.
First, I'll provide the code I used to create the new z values and the dropdown menu, then I'll provide all of the code I used to create these graphs in one chunk (easier to cut and paste...and all that).
This is the data I used for the alternate data in the z field.
import plotly.graph_objects as go
import pandas as pd
import random
z2 = df['GDP (BILLIONS)'] * .667 + 12
random.seed(21)
random.shuffle(z2)
df['z2'] = z2 # example as another column in df
print(df.head()) # validate as expected
z3 = df['GDP (BILLIONS)'] * .2 + 1000
random.seed(231)
random.shuffle(z3) # example as a series outside of df
z4 = df['GDP (BILLIONS)']**(1/3) * df['GDP (BILLIONS)']**(1/2)
random.seed(23)
random.shuffle(z4)
z4 = z4.tolist() # example as a basic Python list
To add buttons to change z, you'll add updatemenus to your layout. Each dict() is a separate dropdown option. At a minimum, each button requires a method, a label, and args. These represent what is changing (method for data, layout, or both), what it's called in the dropdown (label), and the new information (the new z in this example).
args for changes to data (where the method is either restyle or update) can also include the trace the change applies to. So if you had a bar chart and a line graph together, you may have a button that only changes the bar graph.
Using the same structure you have:
updatemenus = [go.layout.Updatemenu(
x = 1, xanchor = 'right', y = 1.15, type = "dropdown",
pad = {'t': 5, 'r': 20, 'b': 5, 'l': 30}, # around all buttons (not indiv buttons)
buttons = list([
dict(
args = [{'z': [df['GDP (BILLIONS)']]}], # original data; nest data in []
label = 'Return to the Original z',
method = 'restyle' # restyle is for trace updates
),
dict(
args = [{'z': [df['z2']]}], # nest data in []
label = 'A different z',
method = 'restyle'
),
dict(
args = [{'z': [z3]}], # nest data in []
label = 'How about this z?',
method = 'restyle'
),
dict(
args = [{'z': [z4]}], # nest data in []
label = 'Last option for z',
method = 'restyle'
)])
)]
All code used to create this graph in one chunk (includes code shown above).
import plotly.graph_objs as go
import pandas as pd
import ssl
import random
# to collect data without an error
ssl._create_default_https_context = ssl._create_unverified_context
# data used in plot
df = pd.read_csv('https://raw.githubusercontent.com/plotly/datasets/master/2014_world_gdp_with_codes.csv')
# z values used in buttons
z2 = df['GDP (BILLIONS)'] * .667 + 12
random.seed(21)
random.shuffle(z2)
df['z2'] = z2 # example as another column in the data frame
print(df.head()) # validate as expected
z3 = df['GDP (BILLIONS)'] * .2 + 1000
random.seed(231)
random.shuffle(z3) # example as a series outside of the data frame
z4 = df['GDP (BILLIONS)']**(1/3) * df['GDP (BILLIONS)']**(1/2)
random.seed(23)
random.shuffle(z4)
z4 = z4.tolist() # example as a basic Python list
data = [go.Choropleth(
locations = df['CODE'], z = df['GDP (BILLIONS)'], text = df['COUNTRY'],
colorscale = [
[0, "rgb(5, 10, 172)"],
[0.35, "rgb(40, 60, 190)"],
[0.5, "rgb(70, 100, 245)"],
[0.6, "rgb(90, 120, 245)"],
[0.7, "rgb(106, 137, 247)"],
[1, "rgb(220, 220, 220)"]],
reversescale = True,
marker = go.choropleth.Marker(
line = go.choropleth.marker.Line(
color = 'rgb(180,180,180)', width = 0.5)),
colorbar = go.choropleth.ColorBar(
tickprefix = '$',
title = 'GDP<br>Billions US$',
len = .6) # I added this for aesthetics
)]
layout = go.Layout(
title = go.layout.Title(text = '2014 Global GDP'),
geo = go.layout.Geo(
showframe = False, showcoastlines = False,
projection = go.layout.geo.Projection(
type = 'equirectangular')
),
annotations = [go.layout.Annotation(
x = 0.55, y = 0.1, xref = 'paper', yref = 'paper',
text = 'Source: <a href="https://www.cia.gov/library/publications/the-world-factbook/fields/2195.html">\
CIA World Factbook</a>',
showarrow = False
)],
updatemenus = [go.layout.Updatemenu(
x = 1, xanchor = 'right', y = 1.15, type = "dropdown",
pad = {'t': 5, 'r': 20, 'b': 5, 'l': 30},
buttons = list([
dict(
args = [{'z': [df['GDP (BILLIONS)']]}], # original data; nest data in []
label = 'Return to the Original z',
method = 'restyle' # restyle is for trace updates only
),
dict(
args = [{'z': [df['z2']]}], # nest data in []
label = 'A different z',
method = 'restyle'
),
dict(
args = [{'z': [z3]}], # nest data in []
label = 'How about this z?',
method = 'restyle'
),
dict(
args = [{'z': [z4]}], # nest data in []
label = 'Last option for z',
method = 'restyle'
)])
)]
)
fig = go.Figure(data = data, layout = layout)
fig.show()
I have three plots based on the same dataset. How can I link all three plots so that when I select a certain species in vbar plot, two scatter plot also change to plot points in that species only.
any help is appreciated~
from bokeh.sampledata.iris import flowers
from bokeh.plotting import figure, output_file, show
from bokeh.models import ColumnDataSource, CategoricalColorMapper
from bokeh.layouts import column, row
#color mapper to color data by species
mapper = CategoricalColorMapper(factors = ['setosa','versicolor', 'virginica'],\
palette = ['green', 'blue', 'red'])
output_file("plots.html")
#group by species and plot barplot for count
species = flowers.groupby('species')
source = ColumnDataSource(species)
p = figure(plot_width = 800, plot_height = 400, title = 'Count by Species', \
x_range = source.data['species'], y_range = (0,60),tools = 'box_select')
p.vbar(x = 'species', top = 'petal_length_count', width = 0.8, source = source,\
nonselection_fill_color = 'gray', nonselection_fill_alpha = 0.2,\
color = {'field': 'species', 'transform': mapper})
labels = LabelSet(x='species', y='petal_length_count', text='petal_length_count',
x_offset=5, y_offset=5, source=source)
p.add_layout(labels)
#scatter plot for sepal length and width
source1 = ColumnDataSource(flowers)
p1 = figure(plot_width = 800, plot_height = 400, tools = 'box_select', title = 'scatter plot for sepal')
p1.circle(x = 'sepal_length', y ='sepal_width', source = source1, \
nonselection_fill_color = 'gray', nonselection_fill_alpha = 0.2, \
color = {'field': 'species', 'transform': mapper})
#scatter plot for petal length and width
p2 = figure(plot_width = 800, plot_height = 400, tools = 'box_select', title = 'scatter plot for petal')
p2.circle(x = 'petal_length', y ='petal_width', source = source1, \
nonselection_fill_color = 'gray', nonselection_fill_alpha = 0.2, \
color = {'field': 'species', 'transform': mapper})
#show all three plots
show(column(p, row(p1, p2)))
I don't think there's some functionality existing for this at the moment. But you can explicitly link two ColumnDataSources with a CustomJS callback:
from bokeh.models import CusomJS
source = ColumnDataSource(species)
source1 = ColumnDataSource(flowers)
source.js_on_change('selected', CustomJS(args=dict(s1=source1), code="""
const indices = cb_obj.selected['1d'].indices;
const species = new Set(indices.map(i => cb_obj.data.species[i]));
s1.selected['1d'].indices = s1.data.species.reduce((acc, s, i) => {if (species.has(s)) acc.push(i); return acc}, []);
s1.select.emit();
"""))
Note that this callback only synchronizes selection from the bar plot to the scatter plots. To make selections on the scatter plots influence the bar plot, you'll have to write some additional code.