Plotly：如何使用 plotly 作为 pandas 的绘图后端来制作不同的绘图？

Question

Setting plotly as a backend for pandas, you can produce plots quickly and easily using:

df.plot()

How can you make this setup produce other plots than lineplots? And what other options are there?

Answer 1

You can define which plot type you'd like to produce through the kind argument in:

df.plot(kind='line')

kind='line' produces the very same plot as in the question. Valid options are:

['scatter', 'line', 'area', 'bar',
 'barh', 'hist', 'box', 'violin',
 'strip', 'funnel', 'density_heatmap',
 'density_contour', 'imshow']

You can easily study them all by running:

for k in kinds[:-1]:
    df.plot(kind=k).show()

Plots:

Notice that I've used kinds[:-1] . This is because imshow is for image data and requires a dataset a bit more complicated than [1,2,3,4,5,6] . Please refer to this and Plotting an image with imshow from a pandas dataframe

In order to set the titles and title colors for all other options, here's a complete code snippet:

import random
import pandas as pd

random.seed(123)
df = pd.DataFrame({'x':[1,2,3,4,5,6]})
pd.options.plotting.backend = "plotly"

kinds = ['scatter', 'line', 'area', 'bar', 'barh', 'hist', 'box', 'violin', 'strip', 'funnel', 'density_heatmap', 'density_contour', 'imshow']

for k in kinds[:-1]:
    fig = df.plot(kind=k, title = k)
    fig.update_layout(title = dict(font=dict(color='#EF553B')))
    fig.show()