xtensor

xtensor copied to clipboard

`xt::pad`: more algorithms & recursion

1

To make `xt::pad` more general and more compliant with NumPy, the xtensor implementation should be extended with: * More (all) algorithms that [NumPy](https://docs.scipy.org/doc/numpy/reference/generated/numpy.pad.html) features. * Recursively call of `pad` if...

tdegeus

Indexing in operator() wrong for xview with xt::newaxis()?

6

I have a small problem with assigning to an xview that was created from an xtensor with `xt::newaxis()`. The code is given below, and also in a separate PR for...

emmenlau

Consider a simple example: we want to take a slice iterator over a sorted array and apply `std::lower_bound` function to it. ``` #include #include #include xarray a = {{{1, 2,...

paceholder

Systematic testing

5

Tests typically use the `xt::xarray`, sometimes `xt::xtensor`, but it turns there are some subtle differences between all available containers/expressions/views that leads to error when switching one for another. A few...

AntoinePrv

performance of `xt::nanstd`

1

Hi, I have been comparing computing the standard deviation myself ("homemade" version in the MWE below) and xtensor's standard deviation: the homemade version seems about 10 times faster than xtensor's...

ThibHlln

xoptional.bool_operation test fails with XSIMD and OpenMP enabled

1

When configured with `-DXTENSOR_USE_{XSIMD,OPENMP}=1 -DBUILD_TESTS=1` the test fails with ``` /tmp/xtensor/test/test_xoptional.cpp:92: TEST CASE: xoptional.bool_operation /tmp/xtensor/test/test_xoptional.cpp:98: ERROR: CHECK_EQ( res(0, 1).value(), true ) is NOT correct! values: CHECK_EQ( {?}, true ) ```...

nodatapoints

The error I get it ``` xpad.hpp:101:25: error: no matching constructor for initialization of 'return_type' (aka 'pyarray') ```

tdegeus

Add `xt::ix_`

2

xtensor is missing cross product indices generation **(numpy.ix_)** https://numpy.org/doc/stable/reference/generated/numpy.ix_.html I am posting this code as a suggestion. It works for my use cases but definitely requires more testing. The idea...

mirmohammad

Give xt::unique() an `axis` argument

1

Rather than flatten the array, it would be cool to find unique rows, or columns, etc. Numpy can currently do this https://www.numpy.org/devdocs/reference/generated/numpy.unique.html#numpy.unique

DavisVaughan

filter missing broadcasting? view missing selector?

3

My eye crossed https://stackoverflow.com/q/74100940/2646505 . The goal there is to mimic ```python import numpy as np keep = np.array([True, False, True]) A = np.array([[1, 2, 3], [4, 5, 6], [7,...

tdegeus