mirror of
https://github.com/hanzoai/ml.git
synced 2026-11-20 23:08:29 +00:00
Three-way per file through the rename: base 39355c6c91, theirs the upstream
merge of main with 35d7ae7c (via 7ef6a694d), ours origin/main.
Lands the barrier thread pool, runtime CPU feature dispatch, aarch64 and x86
repacked quantized kernels (indexed_gemv, gemv_fused_shared_lhs), quantized
embedding on CPU/Metal/CUDA, paged flash-attn kernels, cutile fused MoE, the
in-place op aliasing API, parking_lot storage locks, Metal buffer builders and
debug labels, safetensors 0.8, hf-hub 1.0, tokenizers 0.23.
Ours keeps ROCm, Vulkan, wgpu, the IQ/MXFP4/NVFP4/TQ dtypes, and the i-quant
weight/activation block split in matmul. Our backends ride InplaceOpN and the
quantized embedding; Metal get_rows gains the IQ kernels. ug leaves, as upstream.
Co-authored-by: Hanzo Dev <[email protected]>
hanzo-ml-segformer
- HuggingFace Segformer Model Card
mit-b0- An encoder only pretrained modelsegformer-b0-finetuned-ade-512-512- A fine tuned model for segmentation
How to run the example
If you want you can use the example images from this pull request, download them and supply the path to the image as an argument to the example.
# run the image classification task
cargo run --example segformer classify hanzo-ml-examples/examples/yolo-v8/assets/bike.jpg
# run the segmentation task
cargo run --example segformer segment hanzo-ml-examples/examples/yolo-v8/assets/bike.jpg
Example output for classification:
classification logits [3.275261e-5, 0.0008562019, 0.0008868563, 0.9977506, 0.0002465068, 0.0002241473, 2.846596e-6]
label: hamburger