.. _first_steps_2:

First steps 2: Adding and customizing renderers

In the :ref:previous first steps guide <first_steps_1>, you used Bokeh's |figure| function to render line charts.

In this section, you will use different renderer functions to create various other kinds of graphs. You will also customize what your glyphs look like.

Rendering different glyphs

Bokeh's :ref:plotting <ug_basic> interface supports many different glyphs, such as lines, bars, hex tiles, or other polygons.

.. seealso:: A full list of all supported glyph methods is available in Bokeh's reference guide for the |figure| function. For detailed information on Bokeh's glyphs, see :ref:ug_basic in Bokeh's user guide.

Rendering circles ^^^^^^^^^^^^^^^^^

Use the :func:~bokeh.plotting.figure.scatter function instead of :func:~bokeh.plotting.figure.line to render circles. Circles are the default marker type of the scatter function.

.. code-block:: python

p.scatter(x, y3, legend_label="Objects", color="yellow", size=12)

Replace one of the :func:~bokeh.plotting.figure.line functions in your previous visualization with the :func:~bokeh.plotting.figure.scatter function to create circles.

.. literalinclude:: examples/first_steps_2_add_circles.py :language: python :emphasize-lines: 15

.. bokeh-plot:: docs/first_steps/examples/first_steps_2_add_circles.py :source-position: none

Rendering bars ^^^^^^^^^^^^^^

Similarly, use the :func:~bokeh.plotting.figure.vbar function to render vertical bars.

.. code-block:: python

p.vbar(x=x, top=y2, legend_label="Rate", width=0.5, bottom=0, color="red")

Add the :func:~bokeh.plotting.figure.vbar function to your previous visualization.

.. literalinclude:: examples/first_steps_2_add_bars.py :language: python :emphasize-lines: 14

.. bokeh-plot:: docs/first_steps/examples/first_steps_2_add_bars.py :source-position: none

.. seealso:: To learn more about bar graphs and other ways Bokeh handles categorical data, see :ref:ug_basic_bars in the user guide.

Customizing glyphs

The different renderer functions accept various arguments to control what your glyphs look like.

Defining properties of new glyphs ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

The :func:~bokeh.plotting.figure.scatter function, for example, lets you define aspects like the color or diameter of the circles:

fill_color: the fill color of the circles
fill_alpha: the transparency of the fill color (any value between 0 and 1)
line_color: the fill color of the circles' outlines
size: the size of the circles (in |screen units|)
legend_label: legend entry for the circles

The :func:~bokeh.plotting.figure.scatter function does also support other marker types:

marker: name of the selected form, e.g. "circle", "quad", "y". See :class:~bokeh.core.enums.MarkerType for a complete list.

Note that in the previous examples, you used the color property to define the color of an object. color is an alias that automatically sets all color properties of an object to the same color. For example, passing "yellow" to a circle's color property is the same as setting fill_color and line_color to yellow individually.

In Bokeh, you can :ref:specify colors <ug_styling_colors> in several ways. For example:

Use one of the |named CSS colors| (for example, "firebrick")
Use hexadecimal values, prefaced with a # (for example "#00ff00")
Use a 3-tuple for RGB colors (for example, (100, 100, 255))
Use a 4-tuple for RGBA colors (for example (100, 100, 255, 0.5))

Create circles with the legend label "Objects" and make the circles appear slightly transparent with a red fill color and blue outlines:

.. literalinclude:: examples/first_steps_2_style_circle.py :language: python :emphasize-lines: 11-19

.. bokeh-plot:: docs/first_steps/examples/first_steps_2_style_circle.py :source-position: none

.. _first_steps_2_altering_existing:

Altering properties of existing glyphs ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

If you want to change any property after creating an object, you can define and overwrite the object's attributes directly.

Take the circles from above, for example. You defined the circles to have a red color by passing the argument fill_color="red".

To change the color of your circles from red to blue, you first need to assign a variable name (such as scatter) to the new object when you call the :func:~bokeh.plotting.figure.scatter function.

.. code-block:: python

scatter = p.scatter(
    marker="circle",
    x,
    y,
    size=80,
    legend_label="Objects",
    fill_color="red",
    fill_alpha=0.5,
    line_color="blue",
  )

Next, use that variable to access the object's glyph attribute and change its properties:

.. code-block:: python

glyph = scatter.glyph
glyph.fill_color = "blue"

Generate red circles once more, but this time change their color to blue before outputting the plot:

.. literalinclude:: examples/first_steps_2_style_existing_circle.py :language: python :emphasize-lines: 22-23

.. bokeh-plot:: docs/first_steps/examples/first_steps_2_style_existing_circle.py :source-position: none

.. seealso:: For more information about the various visual properties, see :ref:ug_styling_plots_glyphs and :ref:ug_styling_visuals in the user guide.

Each type of glyph has different properties. Refer to
|figure| in the reference guide to see all available
properties for each glyph method.