A fast, ergonomic and portable tensor library in Nim with a deep learning focus for CPU, GPU and embedded devices via OpenMP, Cuda and OpenCL backends
# What? Fixing a bug with `reshape`, where it mistakes a non-contiguous C layout for contiguous F layout (see [issue #660](https://github.com/mratsim/Arraymancer/issues/660)). # Why A simple illustration why this is a problem: ``` import arraymancer let x = arange[float](0, 6).reshape(2, 3).permute(1, 0) # has shape [3, 2] and is not C-contiguous (but is F-contiguous) let y = x.reshape(3*2).reshape(3, 2) # the reshapes shouldn't change the contents echo x echo y doAssert x == y ``` Right now this assertion fails. # How The check for whether the tensor is F-contiguous is removed. No-copy reshape is only performed if the tensor is C-contiguous.
This issue appears to be discussing a feature request or bug report related to the repository. Based on the content, it seems to be still under discussion. The issue was opened by darkestpigeon and has received 0 comments.