Full Tensor support
lalmei opened this issue · 2 comments
lalmei commented
Current log_image
method can take any np.ndarray
which together with the custom transformation should allow you to log any tensor shape. We would like a smooth experience when a user has a tensor to log. Additionally this use case needs to properly documented
schmidt-jake commented
+1. I have a slightly atypical image dataset that has 6 channels, stored in numpy arrays. I'd love to see whylogs support this kind of image format.
github-actions commented
This issue is stale. Remove stale label or it will be closed tomorrow.