Jan Pipek

Perhaps base it on https://github.com/pydata/sparse

Dtypes for bin contents

3

Support dtypes. To do: - [x] Support for creation in Histogram1D - [x] Support for creation in HistogramND & calculate frequencies - [x] Loss-less arithmetics with constants - [x] Check...

Geospatial histograms

1

Would be nice to be able to plot these in maps. However, this, together with #31, establishes the necessity to keep axes semantics (for axes groups). Not trivial...

And while doing so, extract as much shared code as possible.

Use cupy / numba to speed up computation?

6

Use https://github.com/cupy/cupy as computing backend? Given the API similarity to numpy, it should be trivial. Investigate if numba can lead to any speed up.

Where: https://github.com/janpipek/physt/tree/v05 Requirements: - axis type with abstraction for bin labels etc. - support for units? - at least partial conceptual compatibility with https://github.com/HDembinski/histogram - seamless usability for pandas, xarray,...

And make sure this is reflected both in documentation and identifiers.

Memory efficiency problem

7

When creating histogram from huge data, temporarily huge amount of memory is allocated, though no copy should be created. Suspects: * dropna ??? * weights

Javascript: 'require not defined'