CODEWITHSUNDEEP
pythonpandasnumpymatplotlib
ahajib
ahajib

Reputation: 13510

Plot an array along with multiple pandas columns

I have a pandas dataframe where I'm trying to plot two of its columns for which I do:

from matplotlib import pyplot as plt
import numpy as np    
fig, ax = plt.subplots()
df.plot(x = 'x', y = 'y1', ax = ax, label = 'y1', marker='.')
df.plot(x = 'x', y = 'y2', ax = ax, label = 'y2', marker='.')

The problem happens when I'm trying two plot a third variable (y3) along with those pandas columns. y3 calculated doing the following:

z = np.polyfit(df['x'].values, df['y2'].values, 3)
f = np.poly1d(z)
y3 = f(df['y2'].values)

I have used the following two approached to add this to my previous plot:

ax.plot(x = df['x'].values, y = y3, label = 'estimated', marker = '^')

this does not throw any exception but I can not see the new line added to my plot so basically generates the same plot. I have also tried:

plt.plot(x = df['x'].values, y = y3, label = 'estimated', marker = '^', ax = ax)

which throws:

TypeError: inner() got multiple values for keyword argument 'ax'

How would I add that third line to my plot using values stored in y3 which by the way is a numpy array?

Upvotes: 2

Views: 1909

Answers (1)

Sheldore
Sheldore

Reputation: 39052

For the first two plots, you use your DataFrame df to directly plot the columns using df.plot() where the x and y are required as keyword (optional) arguments (i.e., works without x= and y= as well) (official docs). So using x=... and y=... works for df.plot().

However, in your third plot command, you are using the axis instance ax for plotting using ax.plot() where you are just using the DataFrame values as arguments. ax.plot() accepts x and y values as positional arguments as clarified by ImportanceOfBeingErnest.

So to answer your problem, you need to use 

ax.plot(df['x'].values, y3, label = 'estimated', marker = '^')

by removing x= and y= from your plot command.

The same way, following would work too

plt.plot(df['x'].values, y3, label = 'estimated', marker = '^')

where plt refers to the current axis object.

Upvotes: 2

Related Questions