Problem animating polar plots from measured data

Question:

Problem

I’m trying to animate a polar plot from a measured temperature data from a cylinder using the plotly.express command line_polar by using a dataset of 6 radial values (represented by columns #1 – #6) over 10 rows (represented by column Time) distributed over a polar plot. I’m struggling to make it animate and get the following error:

Error

ValueError: All arguments should have the same length. The length of column argument df[animation_frame] is 10, whereas the length of previously-processed arguments [‘r’, ‘theta’] is 6

According to the help for the parameter "animation_frame" it should be specified as following:

animation_frame (str or int or Series or array-like) – Either a name of a column in data_frame, or a pandas Series or array_like object. Values from this column or array_like are used to assign marks to animation frames.

I’m a bit stumped with this problem since I don’t see why this shouldn’t work, since other use cases seem to use multi-dimensional data with the data with equal rows.

Example of polar plot for t=1

Polar plot

Dataset:

Time #1 #2 #3 #4 #5 #6
1 175 176 179 182 178 173
2 174 175 179 184 178 172
3 175 176 178 183 179 174
4 173 174 178 184 179 174
5 173 174 177 185 180 175
6 173 174 177 185 180 175
7 172 173 176 186 181 176
8 172 173 176 186 181 176
9 171 172 175 187 182 177
10 171 172 175 187 182 177

Code:

import pandas as pd
import plotly.express as px

df = pd.read_excel('TempData.xlsx')

sensor = ["0", "60", "120", "180", "240","300"]
radial_all = ['#1', '#2', '#3', '#4', '#5', '#6']

fig = px.line_polar(df, r=radial_all, theta=sensor, line_close=True,
                     color_discrete_sequence=px.colors.sequential.Plasma_r, template="plotly_dark", animation_frame="Time")
fig.update_polars(radialaxis_range=[160, 190])
fig.update_polars(radialaxis_rangemode="normal")
fig.update_polars(radialaxis=dict(tickvals = [150, 160, 170, 180, 190, 200]))

Thanks in advance!

Asked By: stein

||
Source

Answers:

I have found the solution to this problem, its also possible with scatterpolar but I recommend line_polar from plotly express, its way more elegant and easy. What you need to do is format the data from wide to long format using the pandas command melt(). This will allow you to correctly walk through the data and match it to the animation steps (in this case "Time" column). See following links for helpful info.

Pandas – reshaping-by-melt

pandas.melt()

Resulting code:

import plotly.express as px
import pandas as pd    

df = pd.read_excel('TempData.xlsx')

df_1 = df.melt(id_vars=['Time'], var_name="Sensor", value_name="Temperature",
               value_vars=['#1', '#2', '#3', '#4','#5','#6'])
    
fig = px.line_polar(df_1, r="Temperature", theta="Sensor", line_close=True,
                    line_shape="linear", direction="clockwise",
                   color_discrete_sequence=px.colors.sequential.Plasma_r, template="plotly_dark",
                    animation_frame="Time")
  
fig.show()

Resulting animating plot

Answered By: stein