在python plt.plot中使用一系列作为标记

Question

Is it possible to use a column in a dataframe to scale the marker size in matplotlib? I keep getting an error about using a series when I do the following.

import pandas as pd
import matplotlib.pyplot as plt

my_dict = {'Vx': [16,25,85,45], 'r': [1315,5135,8444,1542], 'ms': [10,50,100, 25]}
df= pd.DataFrame(my_dict)
fig, ax = plt.subplots(1, 1, figsize=(20, 10))
ax.plot(df.Vx, df.r, '.', markersize= df.ms)

when I run

ValueError: setting an array element with a sequence.

I'm guessing it does not like the fact that Im feeding a series to the marker, but there must be a way to make it work...

Answer 1

Better to use the built-in scatter plot function in pandas where you can pass a whole series object as the size param to vary the bubble size:

df.plot.scatter(x=['Vx'], y=['r'], s=df['ms'], c='g')  # df['ms']*5 bubbles more prominent

---

Or, if you want to go via the matplotlib route, you need to pass a scalar value present in the series object each time to the markersize arg.

fig, ax = plt.subplots()
[ax.plot(row['Vx'], row['r'], '.', markersize=row['ms']) for idx, row in df.iterrows()]
plt.show()

Answer 2

Use plt.scatter instead of plt.plot . Scatter lets you specify the size s as well as the color c of the points using a tuple or list.

import pandas as pd
import matplotlib.pyplot as plt

my_dict = {'Vx': [16,25,85,45], 'r': [1315,5135,8444,1542], 'ms': [10,50,100, 25]}
df= pd.DataFrame(my_dict)
fig, ax = plt.subplots(1, 1, figsize=(20, 10))
ax.scatter(df.Vx, df.r, s= df.ms)
plt.show()