Henry Schreiner
Henry Schreiner
parallel-hashmap looks nice and would be easy to add as a dep.
What about adding a "pick" axis bins to Boost.Histogram C++? The current workaround is just that; and that's why it is breaking for boost-histogram with category axis, but it would...
> How about a range of length one? I’m interested in slicing just one category from a category axis. Would that work, or is that even possible already? `h[:, bh.loc("signal")]`...
If we had full UHI support, it would be easy to write a functor that selects categories and use that instead of `bh.sum`. But we don't support it yet.
I can do the OpenMP part, but I don't have enough experience in raw Numpy to know how to write the loops with for's instead of whiles. I automatically try...
I'm sure you are correct. I was mostly focused on development and not on contributions, so I expect some of the docs are too light in that regard. We should...
FYI, `item += 1` stops working if the numbers get large enough for floating point numbers. However, that number is *very* large for doubles.
Also, numpy returns int64s: ```python import numpy as np a,b = np.histogram([1,2,.3]) print(a.dtype) # prints int64 ```
I think this changed in 1.10, at least the 64 part of the int was added then. This is also when numpy got a high-speed 1D histogram algorithm. I wonder...
By the way, the current Numpy code is internally in int64s, and then casts at the end. There's one line that can be dropped to stop the casting. This allows...