Laurent Mazare 89b525b5e7 Convmixer (#1073) ...

* Only optimize float tensors.

* Use full tensors for zeros and ones.

* Add a benchmark for the matmul slowness.

* Add the convmixer model.

* Proper adaptive pooling.

2023-10-11 18:24:32 +01:00

..

generation

Use softmax-last-dim in the quantized example. (#848)

2023-09-14 17:29:24 +01:00

models

Convmixer (#1073)

2023-10-11 18:24:32 +01:00

pipelines

Sketch the candle-transformers crate. (#147)

2023-07-12 13:49:31 +01:00

lib.rs

Move the common quantized-nn code to a shared module. (#1063)

2023-10-09 06:22:22 +01:00

object_detection.rs

Move more models to candle-transformers (#796)

2023-09-10 10:20:18 +01:00

quantized_nn.rs

Move the common quantized-nn code to a shared module. (#1063)

2023-10-09 06:22:22 +01:00

quantized_var_builder.rs

Shared the quantized var-builder code. (#952)

2023-09-24 12:55:07 +01:00

utils.rs

Add some optional repeat penalty. (#623)

2023-08-27 10:48:45 +01:00