Laurent Mazare 9aca398a4f More accelerate optimizations (#427) ...

* Add more tracing to the whisper example.

* Support accelerate in more examples.

* Use accelerate for pointwise functions.

* Use accelerate for binary operations too.

* Bugfix for binary operation: use the rhs before the lhs.

2023-08-13 12:53:34 +01:00

..

accelerate.rs

More accelerate optimizations (#427)

2023-08-13 12:53:34 +01:00

backend.rs

add max_pool2d (#371)

2023-08-09 18:05:26 +01:00

backprop.rs

add max_pool2d (#371)

2023-08-09 18:05:26 +01:00

conv.rs

Add a conv1d benchmark based on the whisper sizes. (#377)

2023-08-09 20:27:03 +01:00

convert.rs

Expose the tensor write-bytes function. (#412)

2023-08-11 17:13:42 +01:00

cpu_backend.rs

Use zero padding in conv1d and conv2d (same as pytorch). (#408)

2023-08-11 14:53:05 +01:00

cpu_kernels.rs

Parallelise the CPU kernels for the conv ops. (#401)

2023-08-11 05:51:58 +01:00

cuda_backend.rs

fix: remove incorrect unwrap (#379)

2023-08-09 21:45:24 +01:00

device.rs

Transpose the weight matrixes for llama2.c. (#321)

2023-08-04 13:32:20 +01:00

display.rs

Remove some dead-code pragmas. (#137)

2023-07-11 09:33:59 +01:00

dtype.rs

Parallelise the CPU kernels for the conv ops. (#401)

2023-08-11 05:51:58 +01:00

dummy_cuda_backend.rs

add max_pool2d (#371)

2023-08-09 18:05:26 +01:00

error.rs

Npy tweaks & error with path (#384)

2023-08-10 06:21:58 +01:00

ggml.rs

Support q5k quantized data. (#320)

2023-08-04 09:51:30 +01:00

indexer.rs

Rename the candle crate to candle-core (#301)

2023-08-02 08:20:22 +01:00

layout.rs

Broadcasting performance optimization (cpu) (#182)

2023-07-17 13:41:09 +01:00

lib.rs

Conv1d optimize (#392)

2023-08-10 15:23:52 +01:00

mkl.rs

Mklize more unary ops. (#191)

2023-07-18 13:32:49 +01:00

npy.rs

Expose the tensor write-bytes function. (#412)

2023-08-11 17:13:42 +01:00

op.rs

More accelerate optimizations (#427)

2023-08-13 12:53:34 +01:00

safetensors.rs

fix: can directly save the loaded weights (#421)

2023-08-12 16:33:29 +01:00

shape.rs

Extract the strides in the conv ops. (#370)

2023-08-09 17:57:05 +01:00

storage.rs

add max_pool2d (#371)

2023-08-09 18:05:26 +01:00

strided_index.rs

Broadcasting performance optimization (cpu) (#182)

2023-07-17 13:41:09 +01:00

tensor.rs

Add a conv1d benchmark based on the whisper sizes. (#377)

2023-08-09 20:27:03 +01:00

utils.rs

Support the Accelerate BLAS on macOS. (#325)

2023-08-05 17:25:24 +01:00

variable.rs

Improve the mnist training example. (#276)

2023-07-29 16:28:22 +01:00