Skip to content
Tutorials Center
My learning

Topics

Lesson 34 of 42 ยท Interactivity and Animation

Animations and Frames

Animate Plotly charts with Plotly Express animation_frame and with go.Frame: play and pause buttons, a frame slider, transitions, fixed axes and bar chart races.

An animation is the easiest way to show change over time: countries moving across a scatter plot as the years pass, a distribution that updates every week, a ball following an orbit. Plotly animations are not video files. They are a list of frames, each frame holding the data (and optionally layout) for one moment, plus a play button and a slider that step through the list. Because it is all in the figure, an animation works in a notebook, a Dash app or an exported HTML page.

Animation with Plotly Express

The shortcut is animation_frame="column". One frame is created for each distinct value of that column, in sorted order. animation_group tells Plotly which rows are "the same object" from frame to frame, so a bubble glides smoothly instead of popping in and out.

Python Gapminder bubbles over time

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Look at what Plotly Express generated: a list of 12 frames (one every five years), a play/pause menu and a slider with 12 steps. These are exactly the updatemenus and sliders you already know, wired to the "animate" method.

Always fix the axis ranges (range_x, range_y), as shown above. Without them, Plotly recomputes the axes in every frame, and the whole picture jumps around, which makes the motion unreadable.

Useful settings:

  • fig.layout.updatemenus[0].buttons[0].args[1]["frame"]["duration"] = 500 changes the time per frame in milliseconds.
  • fig.layout.updatemenus[0].buttons[0].args[1]["transition"]["duration"] = 300 changes the tween time between frames.
  • px.bar(..., animation_frame=..., range_y=[0, max]) animates bars; px.choropleth(..., animation_frame=...) animates a map.

Frames by hand: go.Frame

Behind the scenes a go.Figure has frames, a list of go.Frame objects. Each frame has a name (what the slider points to) and a data list. The data overrides the base figure's traces at the same index. A figure whose base data is the first frame and whose frames list holds the rest is a complete animation once you add the controls.

Python A hand-built animation: a point on a circle

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Two details matter here. traces=[1] in the frame says "this frame only changes trace number 1", so the gray circle is not redrawn. And redraw: False tells Plotly it only needs to update the markers, not rebuild the whole plot, which is much faster. For plots with a heavy layout (maps, 3D) keep redraw on.

The same idea in Plotly.js

In Plotly.js you call Plotly.addFrames (or pass frames to newPlot) and Plotly.animate to play. Here a bar chart "race" reorders bars as the values change.

Bars that grow, frame by frame (Plotly.js)

Press Play, or drag the slider to scrub through time. mode: "immediate" on the slider steps makes the chart jump to that frame without waiting for the previous one to end.

Frames, size and design

flowchart LR
    A["Base figure: data + layout"] --> B["frames: go.Frame list"]
    B --> C["updatemenus: Play / Pause (method animate)"]
    B --> D["sliders: one step per frame (method animate)"]
    C --> E["Browser steps through frames"]
    D --> E

The figure stores every frame; the controls only tell the browser which frame to show next.

  • Size. Each frame stores its data, so a 100-frame animation of 10,000 points is a large page. Animate summaries, not raw data.
  • Meaning. Animation is good for change and engaging stories, but poor for precise comparison across time, because viewers cannot hold earlier frames in memory. Pair it with a static small-multiples view, or a line chart of the key series.
  • Accessibility. Provide the slider so people can pause and step, and avoid fast flashing.
  • Not for real-time. Frames are precomputed. For a stream of new values use a Dash dcc.Interval (see Dash callbacks and live graphs).

Check your understanding

Quick check ยท 4 questions

of 4 answered You scored / 4 ยท perfect!

  1. 1. What does animation_group do in px.scatter?

    Without animation_group, Plotly cannot match points between frames and they appear and disappear instead of gliding.
  2. 2. Why should you set range_x and range_y in an animated chart?

    Autoscaling per frame makes the whole plot shift, which hides the real movement.
  3. 3. What does traces=[1] on a go.Frame mean?

    The frame's data list is applied to the listed trace indexes, leaving other traces untouched.
  4. 4. Which tool is right for a chart that must update with brand-new values every second?

    Frames hold precomputed data. Live data needs a periodic callback that returns a new figure.

Summary

  • Frames are a list of data snapshots; the play button and slider use the animate method.
  • px functions create everything from animation_frame and animation_group.
  • Fix axis ranges, and use traces and redraw=False in hand-built frames for speed.
  • Control timing with frame.duration and transition.duration.
  • Animations show change nicely but are precomputed: use callbacks for live data.

Next: Your First Dash App