Hans Dembinski
Hans Dembinski
As mentioned in #34, we need a way to allow computation from pre-calculated replicates. My original idea was to reuse the existing interface to do this, but documentation and argument...
The jackknife-after-bootstrap method, as described in Efron and Tibshirani's book, is a clever way to compute an uncertainty for a bootstrap estimate, without computing additional replicas. It needs a bit...
I worked out how to iterate the permutation tests until a precision target is reached. It would be great to implement such a functionality also for the functions - `bootstrap.bias`...
One of the first questions I got after the presentation on resample at PyHEP was about parallelization. In principle, resampling methods are perfectly parallelizable, assuming that `fn` is pure (has...
I started to use the library in practice and I found a caveat of our current design. Let's say I want to compute a bias correction *and* the variance of...
### Is your feature request related to a problem? Please describe. I am a particle physicist by trade, and error propagation is very important for us. We need to propagate...
We have a fraction accumulator now, but it does not work correctly for weighted samples. However, we definitely need this in high-energy physics because we often estimate the efficiency of...
Some cost functions can be visualized. To visualize the fitted model, one does ```py m = Minuit(mycost, ...) m.migrad() mycost.visualize(m.values) ``` To simplify this, add a `visualize` method to `Minuit`,...
There are many ways to make mistakes and use the builtin cost functions incorrectly. The user has to provide the model and it is easy to return an array of...
Some people like myself are not interested in the API that requires scipy or algopy. Please make these dependencies optional.