Published November 30, 2018
| Version 0.5.2
Software
Open
scikit-hep/awkward-array: 0.5.2
Authors/Creators
- 1. University of Wisconsin - Madison
- 2. University of Cincinnati
Description
Rewrote JaggedArray._tojagged so that it would not depend on the content being a Numpy array, as part of a fix of #49. (That error was caused by an incompletely filled numpy.empty array in the old JaggedArray._tojagged, but now it's moot.)
Also added performance-testing options. If the following are set:
awkward.array.base.AwkwardArray.allow_tonumpy = False
awkward.array.base.AwkwardArray.allow_iter = False
then no arrays will be auto-convert to Numpy or be iterable in Python (except in __str__ and __repr__). These are the two slowest operations, and refusing them with a RuntimeError may help a user search for bugs.
(The option can also be set on a particular subclass of AwkwardArray or a particular instance.)
Files
scikit-hep/awkward-array-0.5.2.zip
Files
(190.6 kB)
| Name | Size | Download all |
|---|---|---|
|
md5:1a6ca869159b6ea283c034e14e74813e
|
190.6 kB | Preview Download |
Additional details
Related works
- Is supplement to
- https://github.com/scikit-hep/awkward-array/tree/0.5.2 (URL)