Updated to avoid 'PatsyError: model is missing required outcome variables'

[aenfield]

As part of working through chapter 11, I hit an error w/ the following text: 'PatsyError: model is missing required outcome variables', from the GoMining function. I saw this error on both OS X and Windows. I was able to fix the error and get the expected output, on both platforms, by adding a call to 'encode', as shown in the commit associated with this pull request. (I found this idea at https://readditing.com/r/pystats/comments/2cn0go, which refers to this Patsy issue: pydata/patsy#34.)

Two other things:

1. I'm really enjoying the book - thanks for writing it.
2. This is the first time I've submitted a pull request - I tried to follow instructions and it seems like it's worked as it should, but apologies in advance if I've done something incorrectly. If I can fix something to make it work, just let me know.

[AllenDowney]

I made a slightly different change to fix this. Thanks very much for letting me know!

[aenfield]

Great. Thanks again for all of the work on the book (and your Bayes book too).

Andrew

On Mar 4, 2015, at 8:03 AM, Allen Downey [email protected] wrote:

I made a slightly different change to fix this. Thanks very much for letting me know!

—
Reply to this email directly or view it on GitHub.