I recently did a fresh install of Anaconda3 and SMAK on a M1 Mac.
While my previous installation of Anaconda and SMAK proceeded exactly as described on the installation webpage, this time I came across some hurdles.
While I believe that I had installed my previous Anaconda3 for all users, the "/opt/anaconda3" direcoty was in my "User/opt/anaconda3" directory as described in the installation instructions on sams-xrays. Yet this new installation installed in "/opt/anaconda3". i.e. directly on the Macintosh HD, not in the user directory. I may be mistaken about my previous install, but it is worth noting that maybe something changed with the installation defaults. Anyway, this meant that my shortcut no longer worked and I had to edit for the new directory.
The command for creating the smakenv "conda create -n smakenv python==3.10" created the environment in the "osx-arm64" architecture. i.e. the Apple Silicon architecture. The previous architecture is "osx-64".
Not a problem unless you want to install the additional segmentation and image registration modules. When running the "conda install lap -c conda-forge" I got an error: "Packages Not Found Error: The following packages are not available from current channels: -lap ......" This is because this "lap" package does not have a version that works for osx-arm64. Solution: When you create the smakenv, use "conda create --platform osx-64 -n smakenv python==3.10". This creates the smakenv in the architecture that allows you to install the lap package.
The shift to Apple’s M1 and M2 chips under MacOSX Arm architecture introduces new challenges for developers and researchers working with machine learning tools. These chips require recompiled libraries and adjusted workflows to ensure compatibility and performance. MyAssignmenthelp can assist with artificial intelligence assignment help when dealing with such architecture-specific hurdles. Understanding optimization techniques for neural networks on Arm-based systems is crucial, as traditional x86-optimized solutions may not perform similarly. This makes adaptation a key skill in academic tasks involving AI on MacOSX.