How to plot a standard density curve using plotly?

Question

How to plot a density curve in python using plotly?

Specifically, I would like to do that for some standard densities.

---

UPDATE:

So far, the best way of achieving what I ask for (in a Jupyter notebook):

import numpy as np
from scipy import stats

import plotly.graph_objs as go
from plotly.offline import init_notebook_mode, iplot
init_notebook_mode(connected=True)

Plotting

xrange = np.arange(-4.5,5,0.0025)
norm = go.Scatter(x=xrange, 
                  y=stats.norm(loc=0, scale=1).pdf(xrange), 
                  mode='lines',
                  line=dict(width=1.5),
                  name='Normal',
                 )
lognorm = go.Scatter(x=xrange[xrange>0], 
                     y=stats.lognorm(s=1, loc=0, scale=1).pdf(xrange[xrange>0]), 
                     mode='lines',
                     line=dict(width=1.5),
                     name='Lognormal',
                    )
cauchy = go.Scatter(x=xrange, 
                     y=stats.cauchy(loc=0, scale=1).pdf(xrange), 
                     mode='lines',
                     line=dict(width=1.5),
                     name='Cauchy',
                    )
data = [norm, lognorm, cauchy]
iplot(data, show_link=False)

---

I was looking for an easy, parsimonious way, as it is done in Mathematica for instance:

Plot[{PDF[NormalDistribution[], x], PDF[LogNormalDistribution[0, 1], x], PDF[CauchyDistribution[], x]}, {x, -4.5, 4.5}]

Answer 1

The code below uses gaussian_kde. The plots are in iplot(), plotly, as well as matplotlib, and seaborn to compare them together. (Jupyter Notebook 5.0.0, Python 3.6.6)

# Import libraries
import pandas as pd
import numpy as np
from numpy import linspace
import seaborn as sns
from matplotlib import pyplot as plt
import scipy.stats as st
from scipy.stats.kde import gaussian_kde

import plotly.offline as pyo
import plotly.graph_objs as go
from plotly import __version__
%matplotlib inline

import cufflinks as cf
from plotly.offline import download_plotlyjs, init_notebook_mode, plot, iplot 
init_notebook_mode(connected=True)

init_notebook_mode(connected=True)
cf.go_offline()

Create sample data

np.random.seed(1)
data = pd.DataFrame(np.random.randn(500,1), columns=['x'])
data.head(2)

Calculate kde values and put them in a dataframe

df = pd.DataFrame({'x_range': linspace(min(data['x']), max(data['x']), len(data['x'])),
                     'x_kde': kde(x_range) 
                    })

Use iplot() to create plot

df.iplot(x='x_range', y='x_kde')

Use plotly to create plots Note: This plot opens in a new browser window

# Create trace, data and layout 
trace = go.Scatter(x=df['x_range'],y=df['x_kde'], mode='markers',name='markers')
data = [trace]
layout = go.Layout(title='Line Chart')
# Plot figure
fig = go.Figure(data=data, layout=layout)
pyo.offline.plot(fig)

Use Seaborn to plot

sns.distplot(data['x'], hist=False, rug=True)

Use base plot

data['x'].plot.density()

Use Matplotlib to plot

kde = gaussian_kde(data.x)
x_range = linspace(min(data.x), max(data.x), len(data.x))
plt.plot(x_range, kde(x_range) )

Edit . . . . . . . . . . . . . . . . . .

One approach would be to compile all values in a dataframe and then send it to iplot() in one line.

Create dataframe

df = pd.DataFrame({'xrange':np.arange(-4.5,5,0.0025)})
df['norm_x'] = stats.norm(loc=0, scale=1).pdf(df['xrange'])
df['cauchy_x'] = stats.cauchy(loc=0, scale=1).pdf(df['xrange'])
df['lognormal_x'] = stats.lognorm(s=1, loc=0, scale=1).pdf(df['xrange'])
df = pd.DataFrame({'xrange':xrange, 'norm_x':norm_x, 'cauchy_x':cauchy_x, 'lognormal_x':lognormal_x})

Plot using iplot()

df.iplot(x='xrange', y=['norm_x', 'cauchy_x', 'lognormal_x'])