Laurent Mazare 34505fdf3a Avoid using batched-matmul in nn::Linear. (#2883) ...

* Avoid using batched-matmul in nn::Linear.

* Also avoid batched matmul in conv1d.

* Also tweak the conv2d.

* Batched tests.

* Also cover conv2d.

2025-04-12 19:53:58 +02:00

..

conv_tests.rs

Avoid using batched-matmul in nn::Linear. (#2883)

2025-04-12 19:53:58 +02:00

custom_op_tests.rs

Bump the ug dependency. (#2720)

2025-01-16 09:39:16 +01:00

display_tests.rs

…

fortran_tensor_3d.pth

Add support for loading Fortran contiguous tensors (#1672)

2024-02-07 21:49:59 +01:00

grad_tests.rs

Add flip to tensor (#2855)

2025-04-01 09:07:16 +02:00

indexing_tests.rs

Add a basic implementation for slice-assign. (#1377)

2023-11-26 17:31:22 +00:00

layout_tests.rs

Avoid copying the data on squeeze and unsqueeze. (#1884)

2024-03-20 13:04:36 +01:00

matmul_tests.rs

Add a minimal test for the metal bf16 matmul. (#2381)

2024-08-01 11:22:46 +02:00

npy.py

Fix the npy read function and add some testing. (#1080)

2023-10-12 15:25:05 +02:00

pool_tests.rs

Add avg_pool2d metal implementation for the metal backend (#1869)

2024-03-18 18:50:14 +01:00

pth_tests.rs

Add support for loading Fortran contiguous tensors (#1672)

2024-02-07 21:49:59 +01:00

pth.py

Add support for loading Fortran contiguous tensors (#1672)

2024-02-07 21:49:59 +01:00

quantized_tests.rs

upgrade half library to fix rand (#2806)

2025-03-14 09:01:54 +01:00

serialization_tests.rs

Add SliceSafetensors. (#2179)

2024-05-11 13:15:42 +02:00

tensor_tests.rs

Add flip to tensor (#2855)

2025-04-01 09:07:16 +02:00

test_with_key.pt

Enhance pickle to retrieve state_dict with a given key (#1671)

2024-02-06 21:17:33 +01:00

test.npy

Fix the npy read function and add some testing. (#1080)

2023-10-12 15:25:05 +02:00

test.npz

Fix the npy read function and add some testing. (#1080)

2023-10-12 15:25:05 +02:00

test.pt

Fix: pth files don't load on Windows (#1661)

2024-02-06 08:50:55 +01:00