How to plot a line graph with multiple y-values for one x value in plotly?

Question

I want to plot a graph using jupyter noterbook + plotly, based on data that I have grouped by day and ID.

The DataFrame looks like this:

I have tried this in matplotlib but I couldn't figure it out either.

What I want is a graph, with x = hour column and y shows the counts. For example:

Hour 0: x[0] = [2622, 48, 374, 210, 305, 1427, 83, 12]

Hour 1: x[1] = [2920, 25, 357, 140, 283, 79, 14, 53]

... 

with x = [0, 1, 2, ..., 23]

That means each x have many y-values. How can I plot this in plotly? Do I have to do a dimension reduction? If yes, how?

Thanks in advance for help!

vestland · Accepted Answer · 2019-05-22 19:21:32Z

This suggestion will not match your data structure 100%. But we'll handle that later if this turns out to be what you're looking for.

You can achieve what you want by building a plotly figure with multiple traces. If you'd like your datapoints to be a bit jittered to avoid overlapping, you can use go.Box() to get this:

Plot:

Complete code:

This is set up to work in an off-line Jupyter Notebook. Plotly figures are created directly in the notebook.

# imports
import plotly
from plotly import tools
from plotly.offline import download_plotlyjs, init_notebook_mode, plot, iplot
import pandas as pd
import numpy as np
import plotly.plotly as py
import plotly.graph_objs as go

# setup
init_notebook_mode(connected=True)
np.random.seed(123)

# dataframe
df = pd.DataFrame({'x0':[2622, 48, 374, 210, 305, 1427, 83, 12],
                    'x1':[2920, 25, 357, 140, 283, 79, 14, 53]})

# build traces for each x
traces = {}
for col in df.columns:
    traces['trace_' + col] = go.Box(name = col, y=df[col],
                                    boxpoints = 'all',
                                    pointpos = 0,
                                    marker = dict(color = 'rgb(84, 173, 39)'),
                                    line = dict(color = 'rgba(0,0,0,0)'),
                                    fillcolor = 'rgba(0,0,0,0)')
# convert data to form required by plotly
data = list(traces.values())

# build figure
fig = go.Figure(data, layout)

# plot figure
iplot(fig)

Adam · Accepted Answer · 2019-05-22 15:57:22Z

For a given count fix the abscissa
count is a 1D array of length 8
So create 8 identical abscissae and mark their data points
Here it's just about point no need to use plot

Below I just use the two x value you gave, you can use a for loop for the remaining

import numpy as np
import matplotlib.pyplot as plt

y1 = [2622, 48, 374, 210, 305, 1427, 83, 12]
x1 = 0*np.ones(8)

y2 = [2920, 25, 357, 140, 283, 79, 14, 53]
x2 = 1*np.ones(8)

plt.scatter(x1, y1)
plt.scatter(x2, y2)
plt.show()

The corresponding loop code is below

import numpy as np
import matplotlib.pyplot as plt


count = [[2622, 48, 374, 210, 305, 1427, 83, 12], [2920, 25, 357, 140, 283, 79, 14, 53]]
hour = np.ones((8, ) ,dtype = int)

for i in range(2):
  plt.scatter(i*hour, count[i])

plt.show()

Graph

For the entire dataframe just replace

range(2) by range(24)

Collectives™ on Stack Overflow

How to plot a line graph with multiple y-values for one x value in plotly?

2 Answers 2

Comments

Graph

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Graph

Comments

Your Answer

Sign up or log in

Post as a guest

Related