I finally got around to updating some of my crates from ndarray v0.11.* to ndarray v0.13. I noticed that allclose was deprecated in favor of abs_diff_eq. However, I noticed in order to take advantage of it you have to make use of the approx crate within your own and then use one of their exposed macros. I'm not really a fan of this. I'd rather see the traits within approx_array module exposed to users if they have the approx feature added and available within the prelude behind a [cfg(feature = "approx")] flag. The numpy array doc also makes it seem like the abs_diff_eq is available to outside crates but as far as I can tell that's not the case.
I finally got around to updating some of my crates from ndarray v0.11.* to ndarray v0.13. I noticed that
allclosewas deprecated in favor ofabs_diff_eq. However, I noticed in order to take advantage of it you have to make use of theapproxcrate within your own and then use one of their exposed macros. I'm not really a fan of this. I'd rather see the traits within approx_array module exposed to users if they have theapproxfeature added and available within the prelude behind a[cfg(feature = "approx")]flag. Thenumpy array docalso makes it seem like theabs_diff_eqis available to outside crates but as far as I can tell that's not the case.