Seppo Enarvi

I'm thinking about this kind of solution: - Provide a temporary directory as a command line argument. - Store two arrays using `numpy.memmap()`: one that contains the word IDs and...

I also tried to do a 64-bit build, but ended up with even more error messages: gcc -I/sources/atlas-3.10.0//CONFIG/include -g -w -c /sources/atlas-3.10.0//CONFIG/src/atlconf_misc.c gcc -I/sources/atlas-3.10.0//CONFIG/include -g -w -o xconfig /sources/atlas-3.10.0//CONFIG/src/config.c atlconf_misc.o...

I'm using these compilers from the MinGW-W64 project: 32-bit: i686-4.8.2-release-win32-sjlj-rt_v3-rev1 64-bit: x86_64-4.8.2-release-win32-seh-rt_v3-rev1 I also tried specifying the compiler explicitly with --cc flag to make sure it finds the correct compiler.

I ended up using OpenBLAS.

I guess what makes it difficult is that you would need to implement an augmentation that takes multiple images as input. Is that even possible in Albumentations?

> Hi @senarvi, we're currently going through a major, community driven refactor, as outlined in #839. Would you like to work with @luca-medeiros as he's the one who started working...

In the vein of not duplicating code between other Torch codebases, I switched to using the new distance_box_iou() and complete_box_iou() functions. I found out that Torchvision also includes separate iou...

> I'll do a review later but seems that we can mostly improve on docs! How to effectively use the networks with the yolo class etc @luca-medeiros I added a...

This PR also includes the fix to the number of input channels by @kfirgedal . Maybe you could look at this pull request and say what you think?

I added the deeper P6 variant of YOLOv4. The depth of the network and the width of each stage can also be configured freely. I think there are pretty significant...