Section 7.0: Introduction to Plotly's Streaming API¶

Welcome to Plotly's Python API User Guide.

Links to the other sections can be found on the User Guide's homepage

Section 7 is divided, into separate notebooks, as follows:

• 7.0 Streaming API introduction

• 7.1 A first Plotly streaming plot

Quickstart (initialize Plotly figure object and send 1 data point through a stream):

>>> import plotly.plotly as py
>>> from plotly.graph_objs import *
>>> # auto sign-in with credentials or use py.sign_in()
>>> trace1 = Scatter(
        x=[],
        y=[], 
        stream=dict(token='my_stream_id')
    )
>>> data = Data([trace1])
>>> py.plot(data)
>>> s = py.Stream('my_stream_id')
>>> s.open()
>>> s.write(dict(x=1, y=2))
>>> s.close()

Check which version is installed on your machine and please upgrade if needed.

Plotly's Streaming API enables your Plotly plots to update in real-time, without refreshing your browser. In other words, users can continuously send data to Plotly's servers and visualize this data in real-time.

For example, imagine that you have a thermometer (hooked to an Arduino for example) in your attic and you would like to monitor the temperature readings from your laptop. Plotly together with its streaming API makes this project easy to achieve.

With Ploly's Streaming API:

Everyone looking at a Plotly streaming plot sees the same data updating at the same time. Like all Plotly plots, Plotly streaming plots are immediately shareable by shortlink, embedded in website, or in an IPython notebook. Owners of the Plotly plot can edit the plot with the Plotly web GUI and all viewers will see these changes live.

And for the skeptical among us, it's fast: plots update up to 20 times per second.

In this section, we present examples of how to make Plotly streaming plots. Readers looking for info about the nuts and bolts of Plotly's streaming API should refer to this link.

So, we first import a few modules and sign in to Plotly using our credentials file:

What do Plotly streaming plots look like?¶

Data is sent in real-time.
Plotly draws the data in real-time.
Plotly's interactibility happens in real-time.

Get your stream tokens¶

Making Plotly streaming plots requires no modifications to the sign in process; however, users must generate stream tokens or ids. To do so, first sign-in to plot.ly. Once that is done, click on the Settings button in the upper-right corner of the page:

Under the Stream Tokens tab, click on the Generate Token button:

And there you go, you have generated a stream token. Please note that:

You must generate one stream token per trace in your Plotly streaming plot.

If you are looking to run this notebook with you own account, please generate 4 unique stream tokens and add them to your credentials file by entering:

>>> tls.set_credentials_file(stream_ids=[
        "ab4kf5nfdn",
        "kdf5bn4dbn",
        "o72o2p08y5",
        "81dygs4lcy"
    ])

where the stream_ids keyword argument is filled in with your own stream ids.

Note that, in the above, tls.set_credentials() overwrites the existing stream tokens (if any) but does not clear the other keys in your credentials file such as username and api_key.

Once your credentials file is updated with your stream tokens (or stream ids, a synonym), retrieve them as a list:

We are now ready to start making Plotly streaming plots!

The content of this section has been divided into separate IPython notebooks as loading multiple streaming plots at once can cause performance slow downs on some internet connections.

Here are the links to the subsections' notebooks:

• 7.0 Streaming API introduction
• 7.1 A first Plotly streaming plot

In addition, here is a notebook of another Plotly streaming plot:

• Streaming the Poisson Pareto Burst Process by

Emilia Petrisor

Got Questions or Feedback?

Reach us here at: Plotly Community

What's going on at Plotly?