mirror of https://github.com/huggingface/candle.git synced 2025-06-18 11:37:11 +00:00

Files

Nicolas Patry 9c4b4f0da0 Metal quantized modifications proposal.

- Add a device param, wherever needed.
- Create new QMetal storage thing that implements QuantizedType.
- Update everywhere needed.

Fix Python.

Fixing examples.

Fix: fmt + clippy + stub.

Moving everything around.

Only missing the actual implems.

Fixing everything + adding dequantized kernels.

More work.

Fixing matmul.

Fmt + Clippy

Some clippy fixes.

Working state.

Q2K Metal -> Bugged (also present in GGML).
Q4K CPU -> Bugged (present previously, new test catch it).
Q5K CPU -> Bugged (present previously).
Q8_1 Both -> Never really implemented it seems
Q8K metal -> Never implemented in metal

Fixing Q2K bug (present in ggml).

2024-01-15 17:42:58 +01:00

main.rs

Metal quantized modifications proposal.

2024-01-15 17:42:58 +01:00

README.md

Blip attention mask + readme (#1146 )

2023-10-21 22:44:13 +01:00

README.md

candle-blip

The blip-image-captioning model can generate captions for an input image.

Running on an example

cargo run --example blip --release -- --image candle-examples/examples/yolo-v8/assets/bike.jpg

Running on CPU, to run on GPU, build this example with `--features cuda`
loaded image Tensor[dims 3, 384, 384; f32]
model built
several cyclists are riding down a road with cars behind them%