ImportError: cannot import name 'qqplot' from 'seaborn_qqplot' (C:\Python37\lib\site-packages\seaborn_qqplot\__init__.py)

[vinuthna91]

Hi, I tried installing seaborn-qqplot using below command
python3 -m pip install seaborn-qqplot.

Installation was successful.
When I try "import seaborn_qqplot" on python3 terminal, there is no error. However, when I try "from seaborn_qqplot import qqplot", it throws me import error. Kindly help resolve this!

[RonsenbergVI]

Hello,

My apologies, the documentation is not up to date yet. I'll fix that today. In the meantime have a look at the example notebook:

https://github.com/RonsenbergVI/seaborn-qqplot/blob/master/examples/example.ipynb

It explains the new API. I have consolidated all methods in the same method:

from seaborn_qqplot import pplot

the pplot kw argument 'kind' will help you select the type of probability plot you want to show:

• 'q' : quantile plot (i.e. empirical values against empirical quantile function) - univariate
• 'p' : probability plot (i.e. empirical values against cumulative distribution function) - univariate
• 'qq': quantile function against quantile function - bivariate
• 'pp': cumulative distribution function against cumulative distribution function - bivariate

I am also working on a blog post on the topic. I'll add all of these in there when done (hopefully by the end of the week)

If you have any other question do not hesitate, happy to help.

Rene-Jean.

[vinuthna91]

Thanks for the quick reply.
I want to compare against the normal quantiles. Hence, can I use kind="qq" ?

[RonsenbergVI]

Yes do kind='qq'. If you want to compare against a specific distribution you need a scipy.stats.rv_continuous object. The basic code should be:

from seaborn_qqplot import pplot
from scipy.stats import norm


pplot(dataframe, x="your_column", kind='qq', y=norm)

[vinuthna91]

Great, Thanks a lot for the quick help!

[vinuthna91]

Can you give other options that can be passed to pplot?

[RonsenbergVI]

I updated the recommendation, it took a bit longer that expected. I hope it still helps.