Jagrit Digani 9adcd1a650 Support fused masking in Attention (#1924) ...

* Update API to allow mask='causal' in fast::sdpa

* Add fallback

* Update steel::AttnParams

* Fix typo

* WIP, basic causal

* Update tests

* Update benchmarking

* Update masking loop limits

* Add bool masking and update tests

* Update additive mask

* Update benchmarks

* Update benchmarks

* Update tests

* Update for bfloat error

* Update early exit

* Add random seed to tests

2025-03-20 11:01:32 -07:00

..

mlx_tests.py

Make MLX build on x64 macOS (#901)

2024-03-27 06:14:29 -07:00

mpi_test_distributed.py

Support transposed head/seq for kv (#1950)

2025-03-10 10:53:45 -07:00

ring_test_distributed.py

Ring update (#1885)

2025-02-20 14:32:31 -08:00

test_array.py

xor op on arrays (#1875)

2025-02-17 00:24:53 -08:00

test_autograd.py

fix grad with inplace updates (#1961)

2025-03-13 19:13:09 -07:00

test_bf16.py

fix creating array from bf16 tensors in jax / torch (#1305)

2024-08-01 16:20:51 -07:00

test_blas.py

Use same accumulation precision in gemv as gemm (#1962)

2025-03-16 07:13:24 -07:00

test_compile.py

fix cpu compile (#1897)

2025-02-24 14:10:30 -08:00

test_constants.py

Array api (#1289)

2024-07-26 10:40:49 -07:00

test_conv_transpose.py

Fix a couple of slicing bugs (#1827)

2025-02-05 19:50:08 -08:00

test_conv.py

Fix convs by reverting #1803 (#1882)

2025-02-18 14:36:34 -08:00

test_device.py

Adds device context manager (#679)

2024-02-14 14:14:58 -08:00

test_double.py

Double for lapack (#1904)

2025-02-25 11:39:36 -08:00

test_einsum.py

Einsum ellipsis (#1788)

2025-01-25 01:28:03 -08:00

test_eval.py

Fix multistream GPU deadlock (#1969)

2025-03-20 07:19:47 -07:00

test_export_import.py

Cycle leak break (#1856)

2025-02-11 14:45:02 -08:00

test_fast_sdpa.py

Support fused masking in Attention (#1924)

2025-03-20 11:01:32 -07:00

test_fast.py

RMS norm without scaling (#1915)

2025-02-28 20:26:57 -08:00

test_fft.py

throw for invalid case and remove test (#1575)

2024-11-08 12:04:03 -08:00

test_graph.py

Multi output primitives (#330)

2024-01-08 16:39:08 -08:00

test_init.py

feat: Add orthogonal initializer and corresponding tests (#1651)

2025-01-13 07:29:20 -08:00

test_linalg.py

Adds nuclear norm support (#1894)

2025-03-04 13:26:02 -08:00

test_load.py

Limit compile buffers (#1887)

2025-02-19 20:28:13 -08:00

test_losses.py

Improve stability of BCE loss calculation for input probabilities close to or exactly 0 or 1 (#1280)

2024-07-24 08:38:22 -07:00

test_metal.py

Wired (#1510)

2024-10-25 09:35:33 -07:00

test_nn.py

loading empty list is ok when strict = false (#1834)

2025-02-05 16:19:27 -08:00

test_ops.py

fix donation in scan (#1917)

2025-03-03 11:30:59 -08:00

test_optimizers.py

Add a multi optimizer (#1916)

2025-03-04 13:16:35 -08:00

test_quantized.py

Start to cleanup/unify accelerate and common back-ends (Part 1/N) (#1777)

2025-01-29 14:34:49 -08:00

test_random.py

No reshapes in quantized embedding (#1682)

2024-12-09 18:57:38 -08:00

test_reduce.py

Reduce specializations (#1607)

2024-11-21 19:53:00 -08:00

test_tree.py

Add a multi optimizer (#1916)

2025-03-04 13:16:35 -08:00

test_upsample.py

Upsample with bicubic interpolation (#967)

2024-04-10 15:47:22 -07:00

test_vmap.py

fix vmap for flatten (#1955)

2025-03-11 10:42:22 -07:00