I want to create a multi layer graph with the same data frame from pandas.
One should be a boxplot and the other a scatter to see where the company is located.
Is there a way to combine both plots?
boxplot
scatterplot
import pandas as pd
import plotly.express as px
df = pd.read_csv("company_index.csv", sep=";", decimal=",")
print(df)
df_u9 = df.loc[df["company"].isin(["U9"])]
fig_1 = px.box(
df,
x="period",
y="index"
)
fig_2 = px.scatter(
df_u9,
x="period",
y="index"
)
fig_1.show()
fig_2.show()
company_index.csv
period;index;company
1;202,4;U1
1;226,69;U10
1;235,18;U9
1;236,49;U4
1;238,13;U2
1;244,05;U6
1;252,08;U3
1;256,68;U8
1;294,99;U5
1;299,391;U7
2;243,78;U1
2;264,26;U10
2;270,6;U2
2;272,89;U9
2;285,26;U5
2;289,29;U4
2;291,15;U6
2;291,19;U3
2;305,92;U7
2;314,65;U8
3;271,82;U1
3;278,65;U2
3;296,16;U10
3;297,21;U4
3;305,93;U6
3;308,96;U5
3;323,74;U9
3;335,93;U3
3;354,13;U8
3;381,2;U7
4;281,26;U5
4;308,5;U2
4;311,61;U1
4;334,03;U4
4;335,72;U9
4;344,32;U8
4;345,27;U6
4;355,44;U3
4;373,54;U7
4;381,68;U10
5;288,6;U1
5;305,66;U5
5;323,2;U2
5;358,46;U8
5;365,57;U3
5;366,96;U10
5;368,38;U7
5;371,23;U6
5;373,63;U4
5;422,93;U9
6;285,32;U5
6;291,65;U1
6;308,68;U2
6;372,04;U8
6;376,64;U3
6;403,55;U6
6;407,38;U4
6;420,65;U10
6;423,68;U9
6;453,09;U7
Found this solution. Works rather well.
Im still struggling to understand the ".data[0]" but i believe its referring to the first fig in use. Maybe if you have multiple graphs.
import pandas as pd
import plotly.express as px
df = pd.read_csv("company_index.csv", sep=";", decimal=",")
print(df)
df_u9 = df.loc[df["company"].isin(["U9"])].copy()
df_u9["size"] = 1
fig = px.box(
df,
x="period",
y="index"
)
fig.add_trace(px.scatter(
df_u9,
x="period",
y="index",
size="size",
size_max=15,
color_discrete_sequence=(203,153,201)
).data[0])
fig.show()
I want to plot some time series data in plotly where the historic portion of the data has a daily resolution and the data for the current day has minute resolution. Is there a way to somehow "split" the x axis so that for the historic data it only shows the date and for the current data it shows time as well?
Currently it looks like this which is not really that readable
I think the only viable approach would be to put together two subplots. But using the correct setup should make the subplots reach pretty much 100% of what you're describing. You'll only need to adjust a few details like:
fig = make_subplots(rows=1, cols=2,
horizontal_spacing = 0,
shared_yaxes=True,
shared_xaxes=True)
Complete code:
# import pandas as pd
import pandas as pd
import numpy as np
from datetime import datetime, timedelta
from plotly.subplots import make_subplots
import plotly.graph_objects as go
# custom function to set the first
# minute dataset to contiunue from
# the last day in the day dataset
def next_day(date):
s = date
date = datetime.strptime(s, "%Y-%m-%d")
next_date = date + timedelta(days=1)
return(datetime.strftime(next_date, "%Y-%m-%d"))
# data
np.random.seed(10)
n_days = 5
n_minutes = (2*24)
dfd = pd.DataFrame({'time':[t for t in pd.date_range('2020', freq='D', periods=n_days).format()],
'y':np.random.uniform(low=-1, high=1, size=n_days).tolist()})
dfm = pd.DataFrame({'time':[t for t in pd.date_range(next_day(dfd['time'].iloc[-1]), freq='min', periods=n_minutes).format()],
'y':np.random.uniform(low=-1, high=1, size=n_minutes).tolist()})
dfm['y'] = dfm['y'].cumsum()
# subplot setup
fig = make_subplots(rows=1, cols=2,
horizontal_spacing = 0,
shared_yaxes=True,
shared_xaxes=True)
# trace for days
fig.add_trace(
go.Scatter(x=dfd['time'], y=dfd['y'], name = 'days'),
row=1, col=1
)
# trace for minutes
fig.add_trace(
go.Scatter(x=dfm['time'], y=dfm['y'], name = 'minutes'),
row=1, col=2
)
# some x-axis aesthetics
fig.update_layout(xaxis1 = dict(tickangle=0))
fig.update_layout(xaxis2 = dict(tickangle=90))
fig.add_shape( dict(type="line",
x0=dfd['time'].iloc[-1],
y0=dfd['y'].iloc[-1],
x1=dfm['time'].iloc[0],
y1=dfm['y'].iloc[0],
xanchor = 'middle',
xref = 'x1',
yref = 'y1',
line=dict(dash = 'dash',
color="rgba(0,0,255,0.9)",
width=1
)))
fig.update_xaxes(showgrid=False)
fig.update_layout(template = 'plotly_dark')
fig.show()
Using the built in "tips" dataframe in plotly express, I first create a datetime column.
import plotly.express as px
import pandas as pd
from datetime import datetime
df_tips = px.data.tips()
datelist = pd.date_range(datetime.today(), periods=df_tips.shape[0]).tolist()
df_tips['date'] = datelist
Using a column of datetimes as the x-axis gives the error:
px.scatter(df_tips,x='date',y='tip',trendline='ols')
...
TypeError: cannot astype a datetimelike from [datetime64[ns]] to [int32]
Using any other column does not. Is there a good way to do this?
The safest approach is to run the regression on a serialized representation of your dates, and then set th x-axis up to be displayed as strings. By serialized representation I mean, for example, the approach suggested by Ben.T in his comment, or the one used in
Plot best fit line with plotly. Then you can set up the layout of the x-axis using:
fig.update_xaxes(tickangle=45,
tickmode = 'array',
tickvals = df_tips['date'][0::40],
ticktext= [d.strftime('%Y-%m-%d') for d in datelist[0::40]])
The df_tips['date'][0::40] part makes sure there's some space between each tickmark.
Plot 1:
This approach even works well if you'd like to make use of other dimensions of your dataset with, for example: fig = px.scatter(df_tips,x='date',y='tip', color = 'sex', trendline='ols'):
Plot 2:
Complete code:
import plotly.express as px
import pandas as pd
from datetime import datetime
df_tips = px.data.tips()
df_tips['date'] = df_tips.index
datelist = pd.date_range(datetime.today(), periods=df_tips.shape[0]).tolist()
fig = px.scatter(df_tips,x='date',y='tip', trendline='ols')
fig = px.scatter(df_tips,x='date',y='tip', color = 'sex', trendline='ols')
fig.update_xaxes(tickangle=45,
tickmode = 'array',
tickvals = df_tips['date'][0::40],
ticktext= [d.strftime('%Y-%m-%d') for d in datelist])
fig.show()
I want to plot this data to evaluate data availability. I used the following plotting code in Plotly.
import datetime
import plotly.express as px
fig = px.bar(df, x=df.index, y="variable", color='value', orientation="h",
hover_data=[df.index],
height=350,
color_continuous_scale=['firebrick', '#2ca02c'],
title='',
template='plotly_white',
)
The result is just like what I want below.
But, the x-index show numbers. I want a timestamp (month+year) on the x-axis, instead.
Edit
Adding the fllowing
fig.update_layout(yaxis=dict(title=''),
xaxis=dict(
title='Timestamp',
tickformat = '%Y-%b',
)
)
Gives
which seems that the x-axis is not read from the data index.
If you want to use bars it seems to me that you need to find a nice workaround. Have you considered to use Heatmap?
import pandas as pd
import plotly.graph_objs as go
df = pd.read_csv("availability3.txt",
parse_dates=["Timestamp"])\
.drop("Unnamed: 0", axis=1)
# you want to have variable as columns
df = pd.pivot_table(df,
index="Timestamp",
columns="variable",
values="value")
fig = go.Figure()
fig.add_trace(
go.Heatmap(
z=df.values.T,
x=df.index,
y=df.columns,
colorscale='RdYlGn',
xgap=1,
ygap=2)
)
fig.show()
These resources show how to take data from a single Pandas DataFrame and plot different columns subplots on a Plotly graph. I'm interested in creating figures from separate DataFrames and plotting them to the same graph as subplots. Is this possible with Plotly?
https://plot.ly/python/subplots/
https://plot.ly/pandas/subplots/
I'm creating each figure from a dataframe like this:
import pandas as pd
import cufflinks as cf
from plotly.offline import download_plotlyjs, plot,iplot
cf.go_offline()
fig1 = df.iplot(kind='bar',barmode='stack',x='Type',
y=mylist,asFigure=True)
Edit:
Here is an example based on Naren's feedback:
Create the dataframes:
a={'catagory':['loc1','loc2','loc3'],'dogs':[1,5,6],'cats':[3,1,4],'birds':[4,12,2]}
df1 = pd.DataFrame(a)
b={'catagory':['loc1','loc2','loc3'],'dogs':[12,3,5],'cats':[4,6,1],'birds':[7,0,8]}
df2 = pd.DataFrame(b)
The plot will just show the information for the dogs, not the birds or cats:
fig = tls.make_subplots(rows=2, cols=1)
fig1 = df1.iplot(kind='bar',barmode='stack',x='catagory',
y=['dogs','cats','birds'],asFigure=True)
fig.append_trace(fig1['data'][0], 1, 1)
fig2 = df2.iplot(kind='bar',barmode='stack',x='catagory',
y=['dogs','cats','birds'],asFigure=True)
fig.append_trace(fig2['data'][0], 2, 1)
iplot(fig)
Here's a short function in a working example to save a list of figures all to a single HTML file.
def figures_to_html(figs, filename="dashboard.html"):
with open(filename, 'w') as dashboard:
dashboard.write("<html><head></head><body>" + "\n")
for fig in figs:
inner_html = fig.to_html().split('<body>')[1].split('</body>')[0]
dashboard.write(inner_html)
dashboard.write("</body></html>" + "\n")
# Example figures
import plotly.express as px
gapminder = px.data.gapminder().query("country=='Canada'")
fig1 = px.line(gapminder, x="year", y="lifeExp", title='Life expectancy in Canada')
gapminder = px.data.gapminder().query("continent=='Oceania'")
fig2 = px.line(gapminder, x="year", y="lifeExp", color='country')
gapminder = px.data.gapminder().query("continent != 'Asia'")
fig3 = px.line(gapminder, x="year", y="lifeExp", color="continent",
line_group="country", hover_name="country")
figures_to_html([fig1, fig2, fig3])
You can get a dashboard that contains several charts with legends next to each one:
import plotly
import plotly.offline as py
import plotly.graph_objs as go
fichier_html_graphs=open("DASHBOARD.html",'w')
fichier_html_graphs.write("<html><head></head><body>"+"\n")
i=0
while 1:
if i<=40:
i=i+1
#______________________________--Plotly--______________________________________
color1 = '#00bfff'
color2 = '#ff4000'
trace1 = go.Bar(
x = ['2017-09-25','2017-09-26','2017-09-27','2017-09-28','2017-09-29','2017-09-30','2017-10-01'],
y = [25,100,20,7,38,170,200],
name='Debit',
marker=dict(
color=color1
)
)
trace2 = go.Scatter(
x=['2017-09-25','2017-09-26','2017-09-27','2017-09-28','2017-09-29','2017-09-30','2017-10-01'],
y = [3,50,20,7,38,60,100],
name='Taux',
yaxis='y2'
)
data = [trace1, trace2]
layout = go.Layout(
title= ('Chart Number: '+str(i)),
titlefont=dict(
family='Courier New, monospace',
size=15,
color='#7f7f7f'
),
paper_bgcolor='rgba(0,0,0,0)',
plot_bgcolor='rgba(0,0,0,0)',
yaxis=dict(
title='Bandwidth Mbit/s',
titlefont=dict(
color=color1
),
tickfont=dict(
color=color1
)
),
yaxis2=dict(
title='Ratio %',
overlaying='y',
side='right',
titlefont=dict(
color=color2
),
tickfont=dict(
color=color2
)
)
)
fig = go.Figure(data=data, layout=layout)
plotly.offline.plot(fig, filename='Chart_'+str(i)+'.html',auto_open=False)
fichier_html_graphs.write(" <object data=\""+'Chart_'+str(i)+'.html'+"\" width=\"650\" height=\"500\"></object>"+"\n")
else:
break
fichier_html_graphs.write("</body></html>")
print("CHECK YOUR DASHBOARD.html In the current directory")
Result:
You can also try the following using cufflinks:
cf.subplots([df1.figure(kind='bar',categories='category'),
df2.figure(kind='bar',categories='category')],shape=(2,1)).iplot()
And this should give you:
New Answer:
We need to loop through each of the animals and append a new trace to generate what you need. This will give the desired output I am hoping.
import pandas as pd
import numpy as np
import cufflinks as cf
import plotly.tools as tls
from plotly.offline import download_plotlyjs, plot,iplot
cf.go_offline()
import random
def generate_random_color():
r = lambda: random.randint(0,255)
return '#%02X%02X%02X' % (r(),r(),r())
a={'catagory':['loc1','loc2','loc3'],'dogs':[1,5,6],'cats':[3,1,4],'birds':[4,12,2]}
df1 = pd.DataFrame(a)
b={'catagory':['loc1','loc2','loc3'],'dogs':[12,3,5],'cats':[4,6,1],'birds':[7,0,8]}
df2 = pd.DataFrame(b)
#shared Xaxis parameter can make this graph look even better
fig = tls.make_subplots(rows=2, cols=1)
for animal in ['dogs','cats','birds']:
animal_color = generate_random_color()
fig1 = df1.iplot(kind='bar',barmode='stack',x='catagory',
y=animal,asFigure=True,showlegend=False, color = animal_color)
fig.append_trace(fig1['data'][0], 1, 1)
fig2 = df2.iplot(kind='bar',barmode='stack',x='catagory',
y=animal,asFigure=True, showlegend=False, color = animal_color)
#if we do not use the below line there will be two legend
fig2['data'][0]['showlegend'] = False
fig.append_trace(fig2['data'][0], 2, 1)
#additional bonus
#use the below command to use the bar chart three mode
# [stack, overlay, group]
#as shown below
#fig['layout']['barmode'] = 'overlay'
iplot(fig)
Output:
Old Answer:
This will be the solution
Explanation:
Plotly tools has a subplot function to create subplots you should read the documentation for more details here. So I first use cufflinks to create a figure of the bar chart. One thing to note is cufflinks create and object with both data and layout. Plotly will only take one layout parameter as input, hence I take only the data parameter from the cufflinks figure and append_trace it to the make_suplots object. so fig.append_trace() the second parameter is row number and third parameter is column number
import pandas as pd
import cufflinks as cf
import numpy as np
import plotly.tools as tls
from plotly.offline import download_plotlyjs, plot,iplot
cf.go_offline()
fig = tls.make_subplots(rows=2, cols=1)
df = pd.DataFrame(np.random.randint(0,100,size=(100, 4)), columns=list('ABCD'))
fig1 = df.iplot(kind='bar',barmode='stack',x='A',
y='B',asFigure=True)
fig.append_trace(fig1['data'][0], 1, 1)
df2 = pd.DataFrame(np.random.randint(0,100,size=(100, 4)), columns=list('EFGH'))
fig2 = df2.iplot(kind='bar',barmode='stack',x='E',
y='F',asFigure=True)
fig.append_trace(fig2['data'][0], 2, 1)
iplot(fig)
If you want to add a common layout to the subplot I suggest that you do
fig.append_trace(fig2['data'][0], 2, 1)
fig['layout']['showlegend'] = False
iplot(fig)
or even
fig.append_trace(fig2['data'][0], 2, 1)
fig['layout'].update(fig1['layout'])
iplot(fig)
So in the first example before plotting, I access the individual parameters of the layout object and change them, you need to go through layout object properties for refernce.
In the second example before plotting, I update the layout of the figure with the cufflinks generated layout this will produce the same output as we see in cufflinks.
You've already received a few suggestions that work perfectly well. They do however require a lot of coding. Facet / trellis plots using px.bar() will let you produce the plot below using (almost) only this:
px.bar(df, x="category", y="dogs", facet_row="Source")
The only extra steps you'll have to take is to introduce a variable on which to split your data, and then gather or concatenate your dataframes like this:
df1['Source'] = 1
df2['Source'] = 2
df = pd.concat([df1, df2])
And if you'd like to include the other variables as well, just do:
fig = px.bar(df, x="category", y=["dogs", "cats", "birds"], facet_row="Source")
fig.update_layout(barmode = 'group')
Complete code:
# imports
import plotly.express as px
import pandas as pd
# data building
a={'category':['loc1','loc2','loc3'],'dogs':[1,5,6],'cats':[3,1,4],'birds':[4,12,2]}
df1 = pd.DataFrame(a)
b={'category':['loc1','loc2','loc3'],'dogs':[12,3,5],'cats':[4,6,1],'birds':[7,0,8]}
df2 = pd.DataFrame(b)
# data processing
df1['Source'] = 1
df2['Source'] = 2
df = pd.concat([df1, df2])
# plotly figure
fig = px.bar(df, x="category", y="dogs", facet_row="Source")
fig.show()
#fig = px.bar(df, x="category", y=["dogs", "cats", "birds"], facet_row="Source")
#fig.update_layout(barmode = 'group')