Show Sidebar Hide Sidebar

Getting Started with Plotly for Python in Python

Installation and Initialization Steps for Using Chart Studio in Python.

Installation

To install Chart Studio's python package, use the package manager pip inside your terminal.
If you don't have pip installed on your machine, click here for pip's installation instructions.

$ pip install chart_studio
or
$ sudo pip install chart_studio

Plotly's Python package is installed alongside the Chart Studio package and it is updated frequently! To upgrade, run:

$ pip install plotly --upgrade

Initialization for Online Plotting

Chart Studio provides a web-service for hosting graphs! Create a free account to get started. Graphs are saved inside your online Chart Studio account and you control the privacy. Public hosting is free, for private hosting, check out our paid plans.

After installing the Chart Studio package, you're ready to fire up python:

$ python

and set your credentials:

In [1]:
import chart_studio
chart_studio.tools.set_credentials_file(username='DemoAccount', api_key='lr1c37zw81')

You'll need to replace 'DemoAccount' and 'lr1c37zw81' with your Plotly username and API key.
Find your API key here.

The initialization step places a special .plotly/.credentials file in your home directory. Your ~/.plotly/.credentials file should look something like this:

{
    "username": "DemoAccount",
    "stream_ids": ["ylosqsyet5", "h2ct8btk1s", "oxz4fm883b"],
    "api_key": "lr1c37zw81"
}

Online Plot Privacy

Plot can be set to three different type of privacies: public, private or secret.

  • public: Anyone can view this graph. It will appear in your profile and can appear in search engines. You do not need to be logged in to Chart Studio to view this chart.
  • private: Only you can view this plot. It will not appear in the Plotly feed, your profile, or search engines. You must be logged in to Plotly to view this graph. You can privately share this graph with other Chart Studio users in your online Chart Studio account and they will need to be logged in to view this plot.
  • secret: Anyone with this secret link can view this chart. It will not appear in the Chart Studio feed, your profile, or search engines. If it is embedded inside a webpage or an IPython notebook, anybody who is viewing that page will be able to view the graph. You do not need to be logged in to view this plot.

By default all plots are set to public. Users with free account have the permission to keep one private plot. If you need to save private plots, upgrade to a pro account. If you're a Personal or Professional user and would like the default setting for your plots to be private, you can edit your Chart Studio configuration:

In [2]:
import chart_studio
chart_studio.tools.set_config_file(world_readable=False,
                             sharing='private')

For more examples on privacy settings please visit Python privacy documentation

Special Instructions for Chart Studio Enterprise Users

Your API key for account on the public cloud will be different than the API key in Chart Studio Enterprise. Visit https://plotly.your-company.com/settings/api/ to find your Chart Studio Enterprise API key. Remember to replace "your-company.com" with the URL of your Chart Studio Enterprise server. If your company has a Chart Studio Enterprise server, change the Python API endpoint so that it points to your company's Plotly server instead of Plotly's cloud.

In python, enter:

In [3]:
import chart_studio
chart_studio.tools.set_config_file(plotly_domain='https://plotly.your-company.com',
                             plotly_streaming_domain='https://stream-plotly.your-company.com')

Make sure to replace "your-company.com" with the URL of your Chart Studio Enterprise server.

Additionally, you can set your configuration so that you generate private plots by default. For more information on privacy settings see: https://plot.ly/python/privacy/

In python, enter:

In [4]:
import chart_studio
chart_studio.tools.set_config_file(plotly_domain='https://plotly.your-company.com',
                             plotly_streaming_domain='https://stream-plotly.your-company.com',
                             world_readable=False,
                             sharing='private')

Plotly Using virtualenv

Python's virtualenv allows us create multiple working Python environments which can each use different versions of packages. We can use virtualenv from the command line to create an environment using plotly.py version 3.3.0 and a separate one using plotly.py version 2.7.0. See the virtualenv documentation for more info.

Install virtualenv globally
$ sudo pip install virtualenv

Create your virtualenvs
$ mkdir ~/.virtualenvs
$ cd ~/.virtualenvs
$ python -m venv plotly2.7
$ python -m venv plotly3.3

Activate the virtualenv. You will see the name of your virtualenv in parenthesis next to the input promt.
$ source ~/.virtualenvs/plotly2.7/bin/activate
(plotly2.7) $

Install plotly locally to virtualenv (note that we don't use sudo).
(plotly2.7) $ pip install plotly==2.7

Deactivate to exit
(plotly2.7) $ deactivate
$

Jupyter Setup

Install Jupyter into a virtualenv
$ source ~/.virtualenvs/plotly3.3/bin/activate
(plotly3.3) $ pip install notebook

Start the Jupyter kernel from a virtualenv
(plotly3.3) $ jupyter notebook

Start Plotting Online

When plotting online, the plot and data will be saved to your cloud account. There are two methods for plotting online: py.plot() and py.iplot(). Both options create a unique url for the plot and save it in your Plotly account.

  • Use py.plot() to return the unique url and optionally open the url.
  • Use py.iplot() when working in a Jupyter Notebook to display the plot in the notebook.

Copy and paste one of the following examples to create your first hosted Plotly graph using the Plotly Python library:

In [5]:
import chart_studio.plotly as py
import plotly.graph_objects as go

trace0 = go.Scatter(
    x=[1, 2, 3, 4],
    y=[10, 15, 13, 17]
)
trace1 = go.Scatter(
    x=[1, 2, 3, 4],
    y=[16, 5, 11, 9]
)
data = [trace0, trace1]

py.plot(data, filename = 'basic-line', auto_open=True)
Out[5]:
'https://plot.ly/~PythonPlotBot/27/'

Checkout the docstrings for more information:

In [6]:
import chart_studio.plotly as py
help(py.plot)
Help on function plot in module chart_studio.plotly.plotly:

plot(figure_or_data, validate=True, **plot_options)
    Create a unique url for this plot in Plotly and optionally open url.
    
    plot_options keyword arguments:
    filename (string) -- the name that will be associated with this figure
    auto_open (default=True) -- Toggle browser options
        True: open this plot in a new browser tab
        False: do not open plot in the browser, but do return the unique url
    sharing ('public' | 'private' | 'secret') -- Toggle who can view this
                                                  graph
        - 'public': Anyone can view this graph. It will appear in your profile
                    and can appear in search engines. You do not need to be
                    logged in to Plotly to view this chart.
        - 'private': Only you can view this plot. It will not appear in the
                     Plotly feed, your profile, or search engines. You must be
                     logged in to Plotly to view this graph. You can privately
                     share this graph with other Plotly users in your online
                     Plotly account and they will need to be logged in to
                     view this plot.
        - 'secret': Anyone with this secret link can view this chart. It will
                    not appear in the Plotly feed, your profile, or search
                    engines. If it is embedded inside a webpage or an IPython
                    notebook, anybody who is viewing that page will be able to
                    view the graph. You do not need to be logged in to view
                    this plot.
    world_readable (default=True) -- Deprecated: use "sharing".
                                     Make this figure private/public

In [7]:
import chart_studio.plotly as py
import plotly.graph_objects as go

trace0 = go.Scatter(
    x=[1, 2, 3, 4],
    y=[10, 15, 13, 17]
)
trace1 = go.Scatter(
    x=[1, 2, 3, 4],
    y=[16, 5, 11, 9]
)
data = [trace0, trace1]

py.iplot(data, filename = 'basic-line')
Out[7]:

See more examples in our IPython notebook documentation or check out the py.iplot() docstring for more information.

In [8]:
import chart_studio.plotly as py
help(py.iplot)
Help on function iplot in module chart_studio.plotly.plotly:

iplot(figure_or_data, **plot_options)
    Create a unique url for this plot in Plotly and open in IPython.
    
    plot_options keyword arguments:
    filename (string) -- the name that will be associated with this figure
    sharing ('public' | 'private' | 'secret') -- Toggle who can view this graph
        - 'public': Anyone can view this graph. It will appear in your profile
                    and can appear in search engines. You do not need to be
                    logged in to Plotly to view this chart.
        - 'private': Only you can view this plot. It will not appear in the
                     Plotly feed, your profile, or search engines. You must be
                     logged in to Plotly to view this graph. You can privately
                     share this graph with other Plotly users in your online
                     Plotly account and they will need to be logged in to
                     view this plot.
        - 'secret': Anyone with this secret link can view this chart. It will
                    not appear in the Plotly feed, your profile, or search
                    engines. If it is embedded inside a webpage or an IPython
                    notebook, anybody who is viewing that page will be able to
                    view the graph. You do not need to be logged in to view
                    this plot.
    world_readable (default=True) -- Deprecated: use "sharing".
                                     Make this figure private/public

You can also create plotly graphs with matplotlib syntax. Learn more in our matplotlib documentation.

Initialization for Offline Plotting

Plotly allows you to create graphs offline and save them locally. There are also two methods for interactive plotting offline: plotly.io.write_html() and plotly.io.show().

  • Use plotly.io.write_html() to create and standalone HTML that is saved locally and opened inside your web browser.
  • Use plotly.io.show() when working offline in a Jupyter Notebook to display the plot in the notebook.

For information on all of the ways that plotly figures can be displayed, see Displaying plotly figures with plotly for Python.

Copy and paste one of the following examples to create your first offline Plotly graph using the Plotly Python library:

In [9]:
import plotly.graph_objects as go
import plotly.io as pio

fig = go.Figure(go.Scatter(x=[1, 2, 3, 4], y=[4, 3, 2, 1]))
fig.update_layout(title_text='hello world')
pio.write_html(fig, file='hello_world.html', auto_open=True)

Learn more by calling help():

In [10]:
import plotly
help(plotly.io.write_html)
Help on function write_html in module plotly.io._html:

write_html(fig, file, config=None, auto_play=True, include_plotlyjs=True, include_mathjax=False, post_script=None, full_html=True, animation_opts=None, validate=True, default_width='100%', default_height='100%', auto_open=False)
    Write a figure to an HTML file representation
    
    Parameters
    ----------
    fig:
        Figure object or dict representing a figure
    file: str or writeable
        A string representing a local file path or a writeable object
        (e.g. an open file descriptor)
    config: dict or None (default None)
        Plotly.js figure config options
    auto_play: bool (default=True)
        Whether to automatically start the animation sequence on page load
        if the figure contains frames. Has no effect if the figure does not
        contain frames.
    include_plotlyjs: bool or string (default True)
        Specifies how the plotly.js library is included/loaded in the output
        div string.
    
        If True, a script tag containing the plotly.js source code (~3MB)
        is included in the output.  HTML files generated with this option are
        fully self-contained and can be used offline.
    
        If 'cdn', a script tag that references the plotly.js CDN is included
        in the output. HTML files generated with this option are about 3MB
        smaller than those generated with include_plotlyjs=True, but they
        require an active internet connection in order to load the plotly.js
        library.
    
        If 'directory', a script tag is included that references an external
        plotly.min.js bundle that is assumed to reside in the same
        directory as the HTML file. If `file` is a string to a local file path
        and `full_html` is True then
    
        If 'directory', a script tag is included that references an external
        plotly.min.js bundle that is assumed to reside in the same
        directory as the HTML file.  If `file` is a string to a local file
        path and `full_html` is True, then the plotly.min.js bundle is copied
        into the directory of the resulting HTML file. If a file named
        plotly.min.js already exists in the output directory then this file
        is left unmodified and no copy is performed. HTML files generated
        with this option can be used offline, but they require a copy of
        the plotly.min.js bundle in the same directory. This option is
        useful when many figures will be saved as HTML files in the same
        directory because the plotly.js source code will be included only
        once per output directory, rather than once per output file.
    
        If 'require', Plotly.js is loaded using require.js.  This option
        assumes that require.js is globally available and that it has been
        globally configured to know how to find Plotly.js as 'plotly'.
        This option is not advised when full_html=True as it will result
        in a non-functional html file.
    
        If a string that ends in '.js', a script tag is included that
        references the specified path. This approach can be used to point
        the resulting HTML file to an alternative CDN or local bundle.
    
        If False, no script tag referencing plotly.js is included. This is
        useful when the resulting div string will be placed inside an HTML
        document that already loads plotly.js.  This option is not advised
        when full_html=True as it will result in a non-functional html file.
    
    include_mathjax: bool or string (default False)
        Specifies how the MathJax.js library is included in the output html
        div string.  MathJax is required in order to display labels
        with LaTeX typesetting.
    
        If False, no script tag referencing MathJax.js will be included in the
        output.
    
        If 'cdn', a script tag that references a MathJax CDN location will be
        included in the output.  HTML div strings generated with this option
        will be able to display LaTeX typesetting as long as internet access
        is available.
    
        If a string that ends in '.js', a script tag is included that
        references the specified path. This approach can be used to point the
        resulting HTML div string to an alternative CDN.
    post_script: str or list or None (default None)
        JavaScript snippet(s) to be included in the resulting div just after
        plot creation.  The string(s) may include '{plot_id}' placeholders
        that will then be replaced by the `id` of the div element that the
        plotly.js figure is associated with.  One application for this script
        is to install custom plotly.js event handlers.
    full_html: bool (default True)
        If True, produce a string containing a complete HTML document
        starting with an <html> tag.  If False, produce a string containing
        a single <div> element.
    animation_opts: dict or None (default None)
        dict of custom animation parameters to be passed to the function
        Plotly.animate in Plotly.js. See
        https://github.com/plotly/plotly.js/blob/master/src/plots/animation_attributes.js
        for available options. Has no effect if the figure does not contain
        frames, or auto_play is False.
    default_width, default_height: number or str (default '100%')
        The default figure width/height to use if the provided figure does not
        specify its own layout.width/layout.height property.  May be
        specified in pixels as an integer (e.g. 500), or as a css width style
        string (e.g. '500px', '100%').
    validate: bool (default True)
        True if the figure should be validated before being converted to
        JSON, False otherwise.
    auto_open: bool (default True
        If True, open the saved file in a web browser after saving.
        This argument only applies if `full_html` is True.
    Returns
    -------
    str
        Representation of figure as an HTML div string

In [11]:
import plotly.graph_objects as go
import plotly.io as pio

fig = go.Figure(go.Scatter(x=[1, 2, 3, 4], y=[4, 3, 2, 1]))
fig.update_layout(title_text='hello world')
pio.show(fig)