Ethan Cowan

From Theorem 1 (pg. 14) we have a DP guarantee provided that 1. T * q^2 * lambda^2 / sigma^2

I believe this can be merged in with #313

I had the same problem, ~and solved it by using Python 3.5~ Sorry, I meant using torch==1.0.1 instead of the current default (1.5)

There are two avenues here, each with its own set of logical steps: Using DP Count: 1. When the user selects private count = True, then the "create statistic" view...

Needed for computing DP counts: 1. Select any one of the columns in the data set 2. Set a parameter (epsilon/10, etc.) that determines how much budget should be used...

@raprasad Why don't we approach this incrementally, and first build a feature where the user has to answer yes. This way, we can first develop the part of the code...

Another option is to create 2 analysis objects, one for the dp count and one for the rest, and split the budget between them. This way we could reuse the...

@raprasad Starting on this tonight - prototyping a class that will handle SPSS (.sav) and Stata (.dta) files in a single interface and return a Pandas dataframe (based on https://github.com/Roche/pyreadstat)

@raprasad After researching the differences between using pandas and pyreadstat, I'm finding that pandas supports a much smaller subset of file version numbers. The two files I am using for...

@raprasad Currently getting the following error as the variable info page loads: `Failed to open file due to UnicodeDecodeError. ('utf-8' codec can't decode byte 0x96 in position 6: invalid start...