mlops 0.5.0

The meta package: installs every mlops package (mlops-optimizers~=0.2.0, mlops-positional~=0.1.0, mlops-attention~=0.2.0). It has no code of its own.

Install

Install PyTorch and nvfp4effma first, put this wheel and the wheels of the packages (their releases) in one
directory, then:

pip install --no-index -f . mlops

From sources: ./setup.sh all.

mlops-attention 0.2.0: FBLA now covers the whole generation, prompt and decode, on the same compressed cache.
Prefill attends a chunk causally over it in one warp-specialized streaming kernel: producer warps unpack the
records into swizzled shared-memory tiles, consumer warps run QK and PV with an online softmax. Also a complete
text-generation sample (see its release).