Awni Hannun bb1b76d9dc RoPE with frequencies as optional input (#1337) ...

* start rope with freq input

* rope with frequencies

* nits

* fix bug

* fix bug + test

* cleanup

* optional base

2024-08-19 18:30:50 -07:00

..

fft

Feature complete Metal FFT (#1102)

2024-06-06 12:57:25 -07:00

reduction

minor fixes (#1194)

2024-06-12 22:06:49 -07:00

steel

Option to JIT steel gemm / conv (#1139)

2024-05-23 18:07:34 -07:00

arange.h

More jitting (#1132)

2024-05-23 16:23:44 -07:00

arange.metal

More jitting (#1132)

2024-05-23 16:23:44 -07:00

arg_reduce.metal

Option to JIT steel gemm / conv (#1139)

2024-05-23 18:07:34 -07:00

atomic.h

More jitting (#1132)

2024-05-23 16:23:44 -07:00

bf16_math.h

Get metal version from xcode (#1228)

2024-06-26 07:02:11 -07:00

bf16.h

Get metal version from xcode (#1228)

2024-06-26 07:02:11 -07:00

binary_ops.h

JIT compile option for binary minimization (#1091)

2024-05-22 12:57:13 -07:00

binary_two.h

Fixes for large arrays with a few ops (#1299)

2024-07-30 17:18:39 -07:00

binary_two.metal

Fixes for large arrays with a few ops (#1299)

2024-07-30 17:18:39 -07:00

binary.h

Fixes for large arrays with a few ops (#1299)

2024-07-30 17:18:39 -07:00

binary.metal

Fixes for large arrays with a few ops (#1299)

2024-07-30 17:18:39 -07:00

CMakeLists.txt

Add gemv masked to JIT plus some fixes (#1310)

2024-08-07 13:38:07 -07:00

complex.h

feat: metal formatting and pre-commit bump (#1038)

2024-04-30 07:18:09 -07:00

conv.metal

add bfloat conv for windograd (#1306)

2024-08-05 15:51:13 -07:00

copy.h

Fixes for large arrays with a few ops (#1299)

2024-07-30 17:18:39 -07:00

copy.metal

Fixes for large arrays with a few ops (#1299)

2024-07-30 17:18:39 -07:00

defines.h

Add Quantized Ops to the JIT (#1204)

2024-06-12 09:47:12 -07:00

erf.h

JIT compile option for binary minimization (#1091)

2024-05-22 12:57:13 -07:00

expm1f.h

Fix overflow / underflow handling for expm1f (#1278)

2024-07-23 07:29:06 -07:00

fft.h

Feature complete Metal FFT (#1102)

2024-06-06 12:57:25 -07:00

fft.metal

Add Quantized Ops to the JIT (#1204)

2024-06-12 09:47:12 -07:00

gather.h

JIT compile option for binary minimization (#1091)

2024-05-22 12:57:13 -07:00

gemv_masked.h

Add gemv masked to JIT plus some fixes (#1310)

2024-08-07 13:38:07 -07:00

gemv_masked.metal

Add gemv masked to JIT plus some fixes (#1310)

2024-08-07 13:38:07 -07:00

gemv.metal

Masked gemv (#1211)

2024-06-14 09:52:26 -07:00

hadamard.h

Fix bfloat16 Hadamard (#1283)

2024-07-23 14:54:43 -07:00

indexing.h

JIT compile option for binary minimization (#1091)

2024-05-22 12:57:13 -07:00

layer_norm.metal

Fixes for large arrays with a few ops (#1299)

2024-07-30 17:18:39 -07:00

quantized.h

Fused Affine Quantize/Dequantize ops (#1282)

2024-07-29 15:11:38 -07:00

quantized.metal

Fused Affine Quantize/Dequantize ops (#1282)

2024-07-29 15:11:38 -07:00

random.metal

fix rope and random (#1301)

2024-07-31 16:18:25 -07:00

reduce_utils.h

More jitting (#1132)

2024-05-23 16:23:44 -07:00

reduce.h

More jitting (#1132)

2024-05-23 16:23:44 -07:00

reduce.metal

Fix a couple bugs (#1161)

2024-05-28 15:18:18 -07:00

rms_norm.metal

Fixes for large arrays with a few ops (#1299)

2024-07-30 17:18:39 -07:00

rope.metal

RoPE with frequencies as optional input (#1337)

2024-08-19 18:30:50 -07:00

scaled_dot_product_attention_params.h

Metal shaders for memory efficient self attention on large sequences (#964)

2024-06-03 09:16:19 -07:00

scaled_dot_product_attention.metal

Add defines include (#1176)

2024-06-03 09:50:10 -07:00

scan.h

Fix scan (#1188)

2024-06-05 14:21:58 -07:00

scan.metal

More jitting (#1132)

2024-05-23 16:23:44 -07:00

scatter.h

fix scatter + test (#1202)

2024-06-11 14:35:12 -07:00

softmax.h

Fixes for large arrays with a few ops (#1299)

2024-07-30 17:18:39 -07:00

softmax.metal

More jitting (#1132)

2024-05-23 16:23:44 -07:00

sort.h

Fix GPU sort for large arrays (#1285)

2024-07-24 14:37:10 -07:00

sort.metal

Fix strided sort bug (#1236)

2024-06-26 14:32:11 -07:00

ternary_ops.h

JIT compile option for binary minimization (#1091)

2024-05-22 12:57:13 -07:00

ternary.h

Fixes for large arrays with a few ops (#1299)

2024-07-30 17:18:39 -07:00

ternary.metal

Fixes for large arrays with a few ops (#1299)

2024-07-30 17:18:39 -07:00

unary_ops.h

GPU mx.sign for complex64 (#1326)

2024-08-14 07:54:53 -07:00

unary.h

Fixes for large arrays with a few ops (#1299)

2024-07-30 17:18:39 -07:00

unary.metal

GPU mx.sign for complex64 (#1326)

2024-08-14 07:54:53 -07:00

utils.h

More jitting (#1132)

2024-05-23 16:23:44 -07:00