Search code examples
pythonplotly

How to set the color of a plotly line based on a number

I have the following data frame:

import pandas as pd
import numpy as np

df = pd.DataFrame(
    data=np.cumsum( np.sqrt(1 / 1000) * np.random.normal(size=(1000, 10)), axis=0),
    columns=np.array([*range(1, 11)]))

and I want to plot it using plotly express and have the color of each plot to be based on the curves column value and I want the color to continuously evolve. So for example plots 1-3 ish could be yellow, then 4-7 could be orange and 8-10 could be red.

I tried

import plotly.express as px
fig = px.line(df, x=df.index, y=df.columns, colors=df.columns)
fig.show()

but I received the error

All arguments should have the same length. The length of the argument 'color' is 10 where as the length of the previously-processed arguments ['index', '1', '2', '3', '4',..., '10'] is 1000

Essentially what I want to do is translate the following matplot lib code into plotly

import matplotlib.pyplot as plt
import matplotlib.colors as mplc

cmap = plt.get_cmap('viridis', df.shape[1])
norm = mplc.Normalize(vmin=1, vmax=10)
for i, l in enumerate(df.T.values):
    plt.plot(l, color=cmap(norm(i)))
sm = plt.cm.ScalarMappable(cmap=cmap, norm=norm)
plt.colorbar(sm, label='signal number')
plt.show()

which returns the following: enter image description here

Solution

  • Here is the code that I needed. I updated the graph to lines to illustrate that this code will not repeat colors.

      import plotly.graph_objects as go
  import plotly.express as px
  import pandas as pd
  import numpy as np
  import matplotlib.pyplot as plt
  import matplotlib.colors as mplc
  import numpy as np

  no_sigs = 100

  line = lambda x, b: x + b
  lines = np.array(
          [line(time, b) for b in np.linspace(0, 1, 100)]
          )
  line_names = np.array([*range(1, no_sigs + 1)])
  lines_df = pd.DataFrame(
          data=lines.T,
          index=time,
          columns=line_names)

  # rgba colors and plot
  cmap = plt.get_cmap('viridis', no_sigs)
  norm = mplc.Normalize(vmin=sig_names[0], vmax=sig_names[-1])

  # Initialize the color bar
  c_bar_data = np.vstack([lines_df.columns.values,
                          np.empty((2, int(no_sigs)))]).T
  colorbar_df = pd.DataFrame(
          data=c_bar_data,
          columns=['sig_num', 'empt1', 'empt2'])
  fig = px.scatter(colorbar_df,
                   x='empt1',
                   y='empt2',
                   color='sig_num',
                   color_continuous_scale=px.colors.sequential.Viridis)

  # Plot the lines
  for c in lines_df.columns:
      _ = fig.add_trace(
              go.Scatter(
                  y=lines_df[c].values,
                  line={'color': f'rgba{cmap(norm(c))}'},
                  showlegend=False)
              )
  _ = fig.update_xaxes(
          title={'text': 'x-axis'})
  _ = fig.update_yaxes(
          title={'text': 'y-axis'})
  fig.show()

    Here is the output:enter image description here