Is it possible to generate a clock chart using Plotly?

Question

I'm developing a dataviz project and I came across the report generated by Last.FM, in which there is a clock chart to represent the distribution of records by hours.

The chart in question is this:

It is an interactive graph, so I tried to use the Plotly library to try to replicate the chart, but without success.

Is there any way to replicate this in Plotly? Here are the data I need to represent

listeningHour  = df.hour.value_counts().rename_axis('hour').reset_index(name='counts')
listeningHour

   hour counts
0   17  16874
1   18  16703
2   16  14741
3   19  14525
4   23  14440
5   22  13455
6   20  13119
7   21  12766
8   14  11605
9   13  11575
10  15  11491
11  0   10220
12  12  7793
13  1   6057
14  9   3774
15  11  3476
16  10  1674
17  8   1626
18  2   1519
19  3   588
20  6   500
21  7   163
22  4   157
23  5   26

Answer 1

The graph provided by Plotly is a polar bar chart. I have written a code using it with your data. At the time of my research, there does not seem to be a way to place the ticks inside the doughnut. The point of the code is to start at 0:00 in the direction of the angle axis. The clock display is a list of 24 tick places with an empty string and a string every 6 hours. The angle grid is aligned with the center of the bar chart.

import plotly.graph_objects as go

r = df['counts'].tolist()
theta = np.arange(7.5,368,15)
width = [15]*24

ticktexts = [f'$\large{i}$' if i % 6 == 0 else '' for i in np.arange(24)]

fig = go.Figure(go.Barpolar(
    r=r,
    theta=theta,
    width=width,
    marker_color=df['counts'],
    marker_colorscale='Blues',
    marker_line_color="white",
    marker_line_width=2,
    opacity=0.8
))

fig.update_layout(
    template=None,
    polar=dict(
        hole=0.4,
        bgcolor='rgb(223, 223,223)',
        radialaxis=dict(
            showticklabels=False,
            ticks='',
            linewidth=2,
            linecolor='white',
            showgrid=False,
        ),
        angularaxis=dict(
            tickvals=np.arange(0,360,15),
            ticktext=ticktexts,
            showline=True,
            direction='clockwise',
            period=24,
            linecolor='white',
            gridcolor='white',
            showticklabels=True,
            ticks=''
        )
    )
)

fig.show()