Getting ggplot (python) error NameError: name 'stat_function' is not defined

Question

I'm trying to make a plot using ggplot and I'm not sure why I'm getting this error. Here's the code:

from ggplot import *

ggplot(counts, aes(x='pred_prob',y='true_prob',size='count')) + \
    geom_point(color='blue') + \
    stat_function(fun=lambda x: x, color='red') + \
    xlim(-0.05,  1.05) + ylim(-0.05,1.05) 

Here's the error I get:

NameError                                 Traceback (most recent call last)
<ipython-input-71-fdefd49237a1> in <module>()
  2 
  3 baseline = np.mean(is_churn)
----> 4 ggplot(counts, aes(x='pred_prob',y='true_prob',size='count')) + geom_point(color='blue') + stat_function(fun=lambda x: x, color='red') + xlim(-0.05,  1.05) + ylim(-0.05,1.05)
NameError: name 'stat_function' is not defined

I'm not sure why I'm getting this error. Any ideas? I'm using python 3.5.2 and ggplot version 0.11.5

Answer 1

I got the same exact error from the same exact example code

seems the ggplot library had gone through a refactoring.

To install that version use:

pip install -U git+https://github.com/yhat/[email protected]

and then

from ggplot import *
from ggplot.stats.stat_function import stat_function # added line
import pandas as pd
%matplotlib inline

ggplot(pd.DataFrame({'x':[0,500]}), aes(x='x')) + \
    stats.stat_function(fun = lambda x: 50. / (x + 50.)) + \
    ggtitle("Epsilon decay over time") + \
    xlab("amount of feedback") + ylab("epsilon (probability of testing)")

It will produce the same figure as in the post