Commit Graph

  • 4cbe605214
    Feat: Allow per-target Metal debug flags (#2201) Jack Wind 2025-05-20 13:22:26 -0400
  • ab8883dd55
    include mlx::core::version() symbols in the mlx static library (#2207) Clement Liaw 2025-05-20 07:39:11 -0700
  • d5cfa0a2b0 formatting fix Jack 2025-05-19 20:26:52 -0400
  • 53a8cb6a4a
    include mlx::core::version() symbols in the mlx static library Clement Liaw 2025-05-19 16:57:56 -0700
  • eebe73001a
    fix large arg reduce (#2206) Awni Hannun 2025-05-19 13:10:44 -0700
  • bb562bb0fd fix large arg reduce Awni Hannun 2025-05-19 11:46:15 -0700
  • 0359bf02c9
    Nearest upsample (#2202) Angelos Katharopoulos 2025-05-19 11:23:38 -0700
  • d078df1e50 Remove comment from tests Angelos Katharopoulos 2025-05-19 10:58:39 -0700
  • 237f9e58a8
    Fix BEFORE keyword in target_include_directories (#2204) Cheng 2025-05-19 22:10:44 +0900
  • 6d884c6a8e Fix BEFORE keyword in target_include_directories Cheng 2025-05-19 11:11:38 +0000
  • cdd3e9c573 More fixes for nearest Angelos Katharopoulos 2025-05-18 23:41:37 -0700
  • f1d29276b0 Fix Angelos Katharopoulos 2025-05-18 19:43:58 -0700
  • 3cdc4a5f70 Match PT for nearest upsample Angelos Katharopoulos 2025-05-18 19:33:47 -0700
  • c327cdab47 feat: allow per-target Metal debug flags Jack 2025-05-18 19:01:55 -0400
  • 8576e6fe36
    fix conv2d bug + faster conv 1d (#2195) Awni Hannun 2025-05-18 06:05:11 -0700
  • 0654543dcc
    Add complex eigh (#2191) Angelos Katharopoulos 2025-05-18 00:18:43 -0700
  • 2f66115f96 Fix lapack compile on linux Angelos Katharopoulos 2025-05-17 20:58:08 -0700
  • ff1c6fc148 revert sort + flaky test Awni Hannun 2025-05-16 13:40:51 -0700
  • 3b169acf50 fix conv2d bug + faster conv 1d Awni Hannun 2025-05-16 12:12:20 -0700
  • 48ef3e74e2
    reduce vjp for all and any (#2193) Awni Hannun 2025-05-16 08:38:49 -0700
  • 7d4b378952
    Include cuda_bf16.h for bfloat16 overloads (#2192) Cheng 2025-05-16 22:44:42 +0900
  • be41af8954 reduce vjp for all and any Awni Hannun 2025-05-16 06:42:52 -0700
  • 992558dc49 Add NO_GPU_MULTI(Eig) in cuda backend Cheng 2025-05-16 20:39:47 +0900
  • 5e71f2f3ef Include cuda_bf16.h for bfloat16 overloads Cheng 2025-05-11 13:07:15 +0000
  • 7ff5c41e06
    Add set_threadgroup_memory_length to CommandEncoder (#2183) Jack Wind 2025-05-16 03:28:03 -0400
  • 9eb53248ba Add the option for complex eigh Angelos Katharopoulos 2025-05-16 00:05:06 -0700
  • 602f43e3d1
    fix conv grad (#2187) Awni Hannun 2025-05-15 19:20:36 -0700
  • 1a1564f642 index -> int Jack 2025-05-15 22:15:03 -0400
  • a2cadb8218
    real and imag properties (#2189) Awni Hannun 2025-05-15 18:17:50 -0700
  • ae3111cada real and imag properties Awni Hannun 2025-05-15 16:02:31 -0700
  • 48dc69df8e fix conv grad Awni Hannun 2025-05-15 07:13:27 -0700
  • c1eb9d05d9
    non-symmetric eig and eigh (#2188) Awni Hannun 2025-05-15 13:01:44 -0700
  • 64de7c27e1 non-symmetric eig and eigh Awni Hannun 2025-05-15 11:44:05 -0700
  • cf6c939e86
    Fix some complex vjps (#2178) Angelos Katharopoulos 2025-05-14 23:37:12 -0700
  • cf8766e71d Add more tests Angelos Katharopoulos 2025-05-14 23:13:07 -0700
  • 2acf2e003e Comments Angelos Katharopoulos 2025-05-14 21:53:01 -0700
  • 130df35e1b
    Add random normal distribution for complex numbers (#2182) Angelos Katharopoulos 2025-05-13 22:43:45 -0700
  • 4a380d5a80 Cast loc and scale to requested dtype Angelos Katharopoulos 2025-05-13 21:13:44 -0700
  • 0751263dec
    Fix typo in row_reduce_small (#2179) Cheng 2025-05-14 12:19:54 +0900
  • d197c18528 Add set_threadgroup_memory_length to CommandEncoder Jack 2025-05-13 21:45:30 -0400
  • eca2f3eb97
    Add remove_index utility (#2173) Cheng 2025-05-14 09:09:56 +0900
  • 488537e80b Fix the linker errors Angelos Katharopoulos 2025-05-13 15:50:16 -0700
  • 49878758e5 Update the loc and scale to be arrays Angelos Katharopoulos 2025-05-13 15:20:47 -0700
  • c0cac3755c Add a test Angelos Katharopoulos 2025-05-12 23:59:24 -0700
  • 3d93f799df Add standard normal for complex numbers Angelos Katharopoulos 2025-05-12 23:50:38 -0700
  • 3aa9cf3f9e
    Fix put_along_axis for empty arrays (#2181) Angelos Katharopoulos 2025-05-13 14:27:53 -0700
  • 40d2fc1263 Fix shape check Angelos Katharopoulos 2025-05-13 13:21:34 -0700
  • 134ed4a58a Fix put_along_axis for empty arrays Angelos Katharopoulos 2025-05-13 13:12:31 -0700
  • fe8f4c8cf7 Fix typo in row_reduce_small Cheng 2025-05-13 19:48:11 +0900
  • f93cda7a1c Fix the fft tests in C++ Angelos Katharopoulos 2025-05-12 22:52:35 -0700
  • 194f1adbd8 Fixx rfft odd grad and add tests Angelos Katharopoulos 2025-05-12 22:27:12 -0700
  • e1c65e1381 Fix the last fft vjp Angelos Katharopoulos 2025-05-12 17:13:12 -0700
  • 043496bc9b Three ffts done one to go Angelos Katharopoulos 2025-05-12 16:19:12 -0700
  • 23417cee8e Fix complex multiplication derivatives Angelos Katharopoulos 2025-05-12 15:19:54 -0700
  • 8f3d208dce
    Close a couple edge case bugs: hadamard and addmm on empty inputs (#2177) Awni Hannun 2025-05-12 10:48:57 -0700
  • 4e4ec88152 fix Awni Hannun 2025-05-12 10:03:38 -0700
  • bc3357e6ad handle hadamard and addmm on empty inputs Awni Hannun 2025-05-12 09:25:20 -0700
  • caaa3f1f8c
    Small typos in mx.metal deprecations (#2176) Ivan Fioravanti 2025-05-11 15:03:47 +0200
  • b646eb6430 Small typos in mx.metal deprecations ivanfioravanti 2025-05-11 08:44:47 +0200
  • 17bce1c092 Add remove_index utility Cheng 2025-05-11 11:16:48 +0900
  • 3d79254682
    Merge branch 'ml-explore:main' into main Dhruv Srikanth 2025-05-10 16:18:35 +0100
  • aef320bf84 add Muon optimizer Dhruv Srikanth 2025-05-10 16:17:24 +0100
  • 659a51919f
    patch bump (#2162) v0.25.2 Awni Hannun 2025-05-09 14:35:14 -0700
  • 6661387066
    Fix fft for integer overflow (#2161) Awni Hannun 2025-05-09 14:25:12 -0700
  • cd55ab3dac patch bump Awni Hannun 2025-05-09 13:37:45 -0700
  • dff35b2e19 fix fft for integer overflow Awni Hannun 2025-05-09 12:27:01 -0700
  • a7fae8a176
    fix: conv_general differences between gpu, cpu (#2070) ATurker 2025-05-09 20:26:52 +0300
  • 88cc8e0755 add test Awni Hannun 2025-05-09 09:54:20 -0700
  • 76def90b73 fix bugs aturker1 2025-04-29 00:40:13 +0300
  • 7942191a64 fix general_conv padding a-turker 2025-04-12 11:56:10 +0300
  • 83762691ba Fix four step fft fft Angelos Katharopoulos 2025-05-08 14:14:59 -0700
  • 2a41caa00e Add single kernel bluestein Angelos Katharopoulos 2025-05-08 13:15:20 -0700
  • 6593281d25 Refactored four-step Angelos Katharopoulos 2025-05-08 00:25:38 -0700
  • da98e8bce8 Refactored stockham Angelos Katharopoulos 2025-05-06 21:46:21 -0700
  • be57a16a80 More tmp fft changes Angelos Katharopoulos 2025-04-30 22:29:22 -0700
  • 1704809f29 Tmp FFT commit Angelos Katharopoulos 2025-04-30 15:12:39 -0700
  • c8d4d97447 tests added Andrew Sweet 2025-05-07 11:59:54 -0700
  • 072d3791a1
    Update README.md Ankush Goel 2025-05-08 00:20:03 +0530
  • eeaf1fa463
    Update fft_tests.cpp Param Thakkar 2025-05-07 15:24:01 +0530
  • 0cae0bdac8
    CUDA backend: backbone (#2075) Cheng 2025-05-07 13:26:46 +0900
  • 4caeb05c64 Added Symmetric and reflect modes to pad paramthakkar123 2025-05-07 08:15:35 +0530
  • e8ff59451e Add muon optimizer Dhruv Srikanth 2025-05-07 03:29:44 +0100
  • 7c99acb799 split logsumexp split_logsumexp Awni Hannun 2025-05-06 17:10:14 -0700
  • 80002ed42f Updates paramthakkar123 2025-05-06 09:53:10 +0530
  • cbc701f0a7 CUDA backend: backbone Cheng 2025-04-13 16:33:19 +0900
  • 5a1a5d5ed1
    fix input coherent kernel launch (#2153) Awni Hannun 2025-05-05 17:30:50 -0700
  • 09712b8dd0 fix input coherent kernel launch Awni Hannun 2025-05-05 15:52:12 -0700
  • 1683975acf
    Move common gpu primitives to backend/gpu (#2145) Cheng 2025-05-06 05:45:29 +0900
  • af705590ac
    fix batched vector sdpa (#2152) Awni Hannun 2025-05-05 13:13:03 -0700
  • 9762e3ddc0 fix batched vector sdpa Awni Hannun 2025-05-05 12:15:15 -0700
  • 825124af8f
    fix bw for elementwise ops (#2151) Awni Hannun 2025-05-05 06:15:04 -0700
  • 840221e932 fix Awni Hannun 2025-05-03 12:49:00 -0700
  • 9eb5fa764c fix Awni Hannun 2025-05-02 20:40:22 -0700
  • 01a29b51c8 fix Awni Hannun 2025-05-02 20:25:28 -0700
  • 55d6edcaa3 fix Awni Hannun 2025-05-02 17:44:49 -0700
  • d81c2ec3af add compile Awni Hannun 2025-05-02 17:05:04 -0700
  • 57ee5c4954 fix bw for elementwise ops Awni Hannun 2025-05-02 16:47:38 -0700
  • 9c5e7da507
    fix compile merging (#2150) Awni Hannun 2025-05-02 15:08:50 -0700
  • b6d643fc17 fix compile merging Awni Hannun 2025-05-02 13:51:10 -0700
  • a7b13f8584 Move common gpu primitives to backend/gpu Cheng 2025-04-30 16:41:06 +0900