`stan_variable` return pandas dataframe to allow for named columns

[mitzimorris]

Summary:

stan_variable function returns a numpy.nd array, but a pandas dataframe would allow us to keep the column names from the Stan csv file.

Description:

Describe the issue as clearly as possible.

Additional Information:

Provide any additional information here.

Current Version:

[LukasNeugebauer]

Hey!
I haven't ever contributed any code to anything, but I would like to start doing that and this seems like a reasonable first problem.
One question I have is wether the function should return a pd.Series if the variable is a scalar or a Nx1 pd.DataFrame for consistency?
May I have it?

[mitzimorris]

please do!

[mitzimorris]

One question I have is wether the function should return a pd.Series if the variable is a scalar or a Nx1 pd.DataFrame for consistency?

good question - according to SO: https://stackoverflow.com/questions/26047209/what-is-the-difference-between-a-pandas-series-and-a-single-column-dataframe

my instinct is to go with all pd.DataFrames for consistency as long as this doesn't incur performance or memory penalties. here I rely on @ahartikainen's superior Python knowledge - I'm really an old skool Java programmer at heart.

[ahartikainen]

Could you describe a bit the output format? A table with chains stacked in a wide format (wide format == cmdstan output)?

both dataframe and series are fine, but consistent output is better than conditional one.

[LukasNeugebauer]

my instinct is to go with all pd.DataFrames for consistency as long as this doesn't incur performance or memory penalties. here I rely on @ahartikainen's superior Python knowledge - I'm really an old skool Java programmer at heart.

That's also my intuition.

Could you describe a bit the output format? A table with chains stacked in a wide format (wide format == cmdstan output)?

Sure. The output of stan_variable() is a numpy array of post-warmup samples with chains concatenated. So a vector[2] variable with 4 chains of 1000 samples each would become a 4000X2 array. Right now, scalar variables produce arrays without singleton dimension.

both dataframe and series are fine, but consistent output is better than conditional one.

I'll go with DataFrames in both cases.

[mitzimorris]

#287