Edgar Andrés Margffoy Tuay

 I can locally confirm that the tests are passing, the CI error is related to the version of SciPy: https://github.com/scipy/scipy/blob/1d3a067c2ccd0a6efddeb3194163aa9a3879d26e/scipy/interpolate/_ndgriddata.py#L101 I'll skip tests for `scipy

Thanks @asi1024!

@takagi, I've just updated the PR to call `remez` from SciPy, as we discussed with @ev-br on Gitter

@takagi I cannot reproduce any of these errors locally 

Thanks @ev-br for the clarifications! I'll reflect them on the main roadmap. Regarding QHull, I found the following resources: https://www.sciencedirect.com/science/article/abs/pii/S0097849312000350 https://arxiv.org/abs/1508.05488 I haven't read them, but some SO answers point...

@ideasrule, thanks a lot for your contribution! I'll add it to the tracker

Hi @kbauer, thanks for the suggestion! This will be taken into account in the near future when Spyder-Terminal is migrated to allow console program execution from the editor. Right now,...

Sorry for the delay! Here are the benchmark results after [a56161d](https://github.com/cupy/cupy/pull/8328/commits/a56161d3507b39f992e5f9d63e39b02726abceed), this new version matches NumPy implementation in returning the unique output sorted lexicographically in 2D. I tried to add...

Here are the full benchmark results across floating and integer dtypes: ### `dtype=float32` | Size | CuPy (ms) | NumPy (ms) | Speedup | |:----:|:---------:|:----------:|:-------:| | 5 | 0.666010 |...

These are the new benchmark values after [c0fe249](https://github.com/cupy/cupy/pull/8328/commits/c0fe24927abafe244b3f5c074f297afe26e77cc9): ### `dtype=float32` | Size | CuPy (ms) | NumPy (ms) | Speedup | |:----:|:---------:|:----------:|:-------:| | 5 | 0.810840 | 0.069476 | 0.085683...