Peter Quackenbush

ENH: R poly compatibility

16

Hello, I've added the function `poly` which attempts to reproduce R's `poly` function. See here: https://stat.ethz.ch/R-manual/R-devel/library/stats/html/poly.html Thanks.

patsy pickles

18

Hello, This is a continuation effort of #86 and #67. My apologies if this is viewed as stepping on someone else's toes... but this is an interesting--and useful topic--for me....

ENH: Generate var_names from the data and partial predict

5

Hello, I have a proposal that really came about because of the way I've been interacting with patsy. My datasets are kind of long and kind of wide. I have...

joblib.parallel with patsy?

2

So one of my datasets can be super-huge and therefore, super-slow by my impatient standards. I can sit there for 3-5 minutes just waiting for a design matrix to be...

patsy questions/wishlist

4

Hello, I fit some (relatively) large-ish GLMs in statsmodels and have been experimenting with using `patsy` instead of a home rolled thing. My home rolled method isn't very good (I...

WIP: Headbang re-work

4

Hello, If I understand correctly, the headbang triples spatial smoothing is not beign supported because the current one doesn't work too well in python 3. After reviewing, I think I...

Use ast.parse a when breaking lines for the IPython Sphinx Directive

3

Currently, IPython's sphinx directive doesn't really use the `ast.parse` to decide where to break lines, except for functions. This encourages the directive to do just that. This will likely break...

Having a period (.) in a table column name prevents merge and and ffdfdply from working

2

I'm running an ffdfdply operation. Something like this... ``` TrainingData