diff --git a/master/.buildinfo b/master/.buildinfo index 64dc426d9..72ec0780c 100644 --- a/master/.buildinfo +++ b/master/.buildinfo @@ -1,4 +1,4 @@ # Sphinx build info version 1 # This file hashes the configuration used when building these files. When it is not found, a full rebuild will be done. -config: 29a6392189b919d314768a8cb78a5eab +config: b8d6e434bf0cea6b075c4f17e1d5f683 tags: 645f666f9bcd5a90fca523b33c5a78b7 diff --git a/master/.doctrees/environment.pickle b/master/.doctrees/environment.pickle index 852a194e2..33f4448bd 100644 Binary files a/master/.doctrees/environment.pickle and b/master/.doctrees/environment.pickle differ diff --git a/master/.doctrees/getting-started/installation.doctree b/master/.doctrees/getting-started/installation.doctree index 274055c9b..4e8dbde25 100644 Binary files a/master/.doctrees/getting-started/installation.doctree and b/master/.doctrees/getting-started/installation.doctree differ diff --git a/master/.doctrees/getting-started/tutorials/01-vector-add.doctree b/master/.doctrees/getting-started/tutorials/01-vector-add.doctree index 0e773cc17..b0bac0278 100644 Binary files a/master/.doctrees/getting-started/tutorials/01-vector-add.doctree and b/master/.doctrees/getting-started/tutorials/01-vector-add.doctree differ diff --git a/master/.doctrees/getting-started/tutorials/02-fused-softmax.doctree b/master/.doctrees/getting-started/tutorials/02-fused-softmax.doctree index 2dd5f5ee3..43aae2e65 100644 Binary files a/master/.doctrees/getting-started/tutorials/02-fused-softmax.doctree and b/master/.doctrees/getting-started/tutorials/02-fused-softmax.doctree differ diff --git a/master/.doctrees/getting-started/tutorials/03-matrix-multiplication.doctree b/master/.doctrees/getting-started/tutorials/03-matrix-multiplication.doctree index e10b8ad9f..17ba33c20 100644 Binary files a/master/.doctrees/getting-started/tutorials/03-matrix-multiplication.doctree and b/master/.doctrees/getting-started/tutorials/03-matrix-multiplication.doctree differ diff --git a/master/.doctrees/getting-started/tutorials/04-low-memory-dropout.doctree b/master/.doctrees/getting-started/tutorials/04-low-memory-dropout.doctree index 9661ce0ff..83b307eb8 100644 Binary files a/master/.doctrees/getting-started/tutorials/04-low-memory-dropout.doctree and b/master/.doctrees/getting-started/tutorials/04-low-memory-dropout.doctree differ diff --git a/master/.doctrees/getting-started/tutorials/05-layer-norm.doctree b/master/.doctrees/getting-started/tutorials/05-layer-norm.doctree index 7c4d5ebef..cd1168bb8 100644 Binary files a/master/.doctrees/getting-started/tutorials/05-layer-norm.doctree and b/master/.doctrees/getting-started/tutorials/05-layer-norm.doctree differ diff --git a/master/.doctrees/getting-started/tutorials/index.doctree b/master/.doctrees/getting-started/tutorials/index.doctree index 60cfda12c..1a9951b93 100644 Binary files a/master/.doctrees/getting-started/tutorials/index.doctree and b/master/.doctrees/getting-started/tutorials/index.doctree differ diff --git a/master/.doctrees/getting-started/tutorials/sg_execution_times.doctree b/master/.doctrees/getting-started/tutorials/sg_execution_times.doctree index d10eb4b35..ea2358f7d 100644 Binary files a/master/.doctrees/getting-started/tutorials/sg_execution_times.doctree and b/master/.doctrees/getting-started/tutorials/sg_execution_times.doctree differ diff --git a/master/.doctrees/index.doctree b/master/.doctrees/index.doctree index 09afbce0a..25b617730 100644 Binary files a/master/.doctrees/index.doctree and b/master/.doctrees/index.doctree differ diff --git a/master/.doctrees/programming-guide/chapter-1/introduction.doctree b/master/.doctrees/programming-guide/chapter-1/introduction.doctree index 80450577e..315edd43c 100644 Binary files a/master/.doctrees/programming-guide/chapter-1/introduction.doctree and b/master/.doctrees/programming-guide/chapter-1/introduction.doctree differ diff --git a/master/.doctrees/programming-guide/chapter-2/related-work.doctree b/master/.doctrees/programming-guide/chapter-2/related-work.doctree index 4b08a31e9..84f922b20 100644 Binary files a/master/.doctrees/programming-guide/chapter-2/related-work.doctree and b/master/.doctrees/programming-guide/chapter-2/related-work.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.Config.doctree b/master/.doctrees/python-api/generated/triton.Config.doctree index d1a286e7f..f7eb29acd 100644 Binary files a/master/.doctrees/python-api/generated/triton.Config.doctree and b/master/.doctrees/python-api/generated/triton.Config.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.autotune.doctree b/master/.doctrees/python-api/generated/triton.autotune.doctree index cea07328f..e7f46c00b 100644 Binary files a/master/.doctrees/python-api/generated/triton.autotune.doctree and b/master/.doctrees/python-api/generated/triton.autotune.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.heuristics.doctree b/master/.doctrees/python-api/generated/triton.heuristics.doctree index 367ac0e35..a78e4de13 100644 Binary files a/master/.doctrees/python-api/generated/triton.heuristics.doctree and b/master/.doctrees/python-api/generated/triton.heuristics.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.jit.doctree b/master/.doctrees/python-api/generated/triton.jit.doctree index 36e55d4ed..30f4fcc6c 100644 Binary files a/master/.doctrees/python-api/generated/triton.jit.doctree and b/master/.doctrees/python-api/generated/triton.jit.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.arange.doctree b/master/.doctrees/python-api/generated/triton.language.arange.doctree index 630aa67a9..92a01e204 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.arange.doctree and b/master/.doctrees/python-api/generated/triton.language.arange.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.atomic_add.doctree b/master/.doctrees/python-api/generated/triton.language.atomic_add.doctree index 1c035d663..8aaf4f591 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.atomic_add.doctree and b/master/.doctrees/python-api/generated/triton.language.atomic_add.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.atomic_cas.doctree b/master/.doctrees/python-api/generated/triton.language.atomic_cas.doctree index 5493d105a..ef3a4effa 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.atomic_cas.doctree and b/master/.doctrees/python-api/generated/triton.language.atomic_cas.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.atomic_max.doctree b/master/.doctrees/python-api/generated/triton.language.atomic_max.doctree index eb28751e5..9ffa73716 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.atomic_max.doctree and b/master/.doctrees/python-api/generated/triton.language.atomic_max.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.atomic_min.doctree b/master/.doctrees/python-api/generated/triton.language.atomic_min.doctree index 89225099e..75ccfeeeb 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.atomic_min.doctree and b/master/.doctrees/python-api/generated/triton.language.atomic_min.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.atomic_xchg.doctree b/master/.doctrees/python-api/generated/triton.language.atomic_xchg.doctree index cd4c0ad88..c41023337 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.atomic_xchg.doctree and b/master/.doctrees/python-api/generated/triton.language.atomic_xchg.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.broadcast_to.doctree b/master/.doctrees/python-api/generated/triton.language.broadcast_to.doctree index 76293ec5f..93313e906 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.broadcast_to.doctree and b/master/.doctrees/python-api/generated/triton.language.broadcast_to.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.cos.doctree b/master/.doctrees/python-api/generated/triton.language.cos.doctree index 5e1558eb2..30b01cea0 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.cos.doctree and b/master/.doctrees/python-api/generated/triton.language.cos.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.dot.doctree b/master/.doctrees/python-api/generated/triton.language.dot.doctree index 2da90a6dd..978b6cfd5 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.dot.doctree and b/master/.doctrees/python-api/generated/triton.language.dot.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.exp.doctree b/master/.doctrees/python-api/generated/triton.language.exp.doctree index 3aaa48adf..4810bf987 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.exp.doctree and b/master/.doctrees/python-api/generated/triton.language.exp.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.load.doctree b/master/.doctrees/python-api/generated/triton.language.load.doctree index 1e6b05d7b..9916a09eb 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.load.doctree and b/master/.doctrees/python-api/generated/triton.language.load.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.log.doctree b/master/.doctrees/python-api/generated/triton.language.log.doctree index bd29bd539..fb1aaf6c8 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.log.doctree and b/master/.doctrees/python-api/generated/triton.language.log.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.max.doctree b/master/.doctrees/python-api/generated/triton.language.max.doctree index a91558835..2ad96c4a5 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.max.doctree and b/master/.doctrees/python-api/generated/triton.language.max.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.maximum.doctree b/master/.doctrees/python-api/generated/triton.language.maximum.doctree index a9645824b..61e96b61b 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.maximum.doctree and b/master/.doctrees/python-api/generated/triton.language.maximum.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.min.doctree b/master/.doctrees/python-api/generated/triton.language.min.doctree index 0f044e916..3c41e2b18 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.min.doctree and b/master/.doctrees/python-api/generated/triton.language.min.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.minimum.doctree b/master/.doctrees/python-api/generated/triton.language.minimum.doctree index 3e0188f19..6056221df 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.minimum.doctree and b/master/.doctrees/python-api/generated/triton.language.minimum.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.multiple_of.doctree b/master/.doctrees/python-api/generated/triton.language.multiple_of.doctree index 9ad9b44c4..58458832d 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.multiple_of.doctree and b/master/.doctrees/python-api/generated/triton.language.multiple_of.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.num_programs.doctree b/master/.doctrees/python-api/generated/triton.language.num_programs.doctree index 6644c7742..41a8e0126 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.num_programs.doctree and b/master/.doctrees/python-api/generated/triton.language.num_programs.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.program_id.doctree b/master/.doctrees/python-api/generated/triton.language.program_id.doctree index bf30ce67a..336e6c664 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.program_id.doctree and b/master/.doctrees/python-api/generated/triton.language.program_id.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.rand.doctree b/master/.doctrees/python-api/generated/triton.language.rand.doctree index 946a0f177..496c4240f 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.rand.doctree and b/master/.doctrees/python-api/generated/triton.language.rand.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.randint.doctree b/master/.doctrees/python-api/generated/triton.language.randint.doctree index 0f7e5f95e..148cc9d10 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.randint.doctree and b/master/.doctrees/python-api/generated/triton.language.randint.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.randint4x.doctree b/master/.doctrees/python-api/generated/triton.language.randint4x.doctree index 5fc876148..0ac6badd0 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.randint4x.doctree and b/master/.doctrees/python-api/generated/triton.language.randint4x.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.randn.doctree b/master/.doctrees/python-api/generated/triton.language.randn.doctree index c208908ce..af2776a50 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.randn.doctree and b/master/.doctrees/python-api/generated/triton.language.randn.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.ravel.doctree b/master/.doctrees/python-api/generated/triton.language.ravel.doctree index 465bb35ec..195deed72 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.ravel.doctree and b/master/.doctrees/python-api/generated/triton.language.ravel.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.reshape.doctree b/master/.doctrees/python-api/generated/triton.language.reshape.doctree index 5579f869f..4f6877a32 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.reshape.doctree and b/master/.doctrees/python-api/generated/triton.language.reshape.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.sigmoid.doctree b/master/.doctrees/python-api/generated/triton.language.sigmoid.doctree index c59d3bcd6..7dc52f4b9 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.sigmoid.doctree and b/master/.doctrees/python-api/generated/triton.language.sigmoid.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.sin.doctree b/master/.doctrees/python-api/generated/triton.language.sin.doctree index d4b885b8d..728882a1a 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.sin.doctree and b/master/.doctrees/python-api/generated/triton.language.sin.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.softmax.doctree b/master/.doctrees/python-api/generated/triton.language.softmax.doctree index eb2b73ddd..5f56e0358 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.softmax.doctree and b/master/.doctrees/python-api/generated/triton.language.softmax.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.sqrt.doctree b/master/.doctrees/python-api/generated/triton.language.sqrt.doctree index a020205f3..2ba75d669 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.sqrt.doctree and b/master/.doctrees/python-api/generated/triton.language.sqrt.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.store.doctree b/master/.doctrees/python-api/generated/triton.language.store.doctree index a876c7651..9cac4421d 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.store.doctree and b/master/.doctrees/python-api/generated/triton.language.store.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.sum.doctree b/master/.doctrees/python-api/generated/triton.language.sum.doctree index d63942976..ac764a42b 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.sum.doctree and b/master/.doctrees/python-api/generated/triton.language.sum.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.where.doctree b/master/.doctrees/python-api/generated/triton.language.where.doctree index 5db4e9bf0..5007b347d 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.where.doctree and b/master/.doctrees/python-api/generated/triton.language.where.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.language.zeros.doctree b/master/.doctrees/python-api/generated/triton.language.zeros.doctree index bc27112dd..490c97888 100644 Binary files a/master/.doctrees/python-api/generated/triton.language.zeros.doctree and b/master/.doctrees/python-api/generated/triton.language.zeros.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.testing.Benchmark.doctree b/master/.doctrees/python-api/generated/triton.testing.Benchmark.doctree index 3d7b9e58a..79f20c11c 100644 Binary files a/master/.doctrees/python-api/generated/triton.testing.Benchmark.doctree and b/master/.doctrees/python-api/generated/triton.testing.Benchmark.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.testing.do_bench.doctree b/master/.doctrees/python-api/generated/triton.testing.do_bench.doctree index ca5c61d6b..fd91dae96 100644 Binary files a/master/.doctrees/python-api/generated/triton.testing.do_bench.doctree and b/master/.doctrees/python-api/generated/triton.testing.do_bench.doctree differ diff --git a/master/.doctrees/python-api/generated/triton.testing.perf_report.doctree b/master/.doctrees/python-api/generated/triton.testing.perf_report.doctree index d21e1ea19..247f09af9 100644 Binary files a/master/.doctrees/python-api/generated/triton.testing.perf_report.doctree and b/master/.doctrees/python-api/generated/triton.testing.perf_report.doctree differ diff --git a/master/.doctrees/python-api/triton.doctree b/master/.doctrees/python-api/triton.doctree index accc0082c..db0f4c5ae 100644 Binary files a/master/.doctrees/python-api/triton.doctree and b/master/.doctrees/python-api/triton.doctree differ diff --git a/master/.doctrees/python-api/triton.language.doctree b/master/.doctrees/python-api/triton.language.doctree index 3abd27b73..a47d188f5 100644 Binary files a/master/.doctrees/python-api/triton.language.doctree and b/master/.doctrees/python-api/triton.language.doctree differ diff --git a/master/.doctrees/python-api/triton.testing.doctree b/master/.doctrees/python-api/triton.testing.doctree index 9594bef18..04558c930 100644 Binary files a/master/.doctrees/python-api/triton.testing.doctree and b/master/.doctrees/python-api/triton.testing.doctree differ diff --git a/master/_downloads/662999063954282841dc90b8945f85ce/tutorials_jupyter.zip b/master/_downloads/662999063954282841dc90b8945f85ce/tutorials_jupyter.zip index cfea68960..2563b5a09 100644 Binary files a/master/_downloads/662999063954282841dc90b8945f85ce/tutorials_jupyter.zip and b/master/_downloads/662999063954282841dc90b8945f85ce/tutorials_jupyter.zip differ diff --git a/master/_downloads/763344228ae6bc253ed1a6cf586aa30d/tutorials_python.zip b/master/_downloads/763344228ae6bc253ed1a6cf586aa30d/tutorials_python.zip index 9731194a9..9944c8df4 100644 Binary files a/master/_downloads/763344228ae6bc253ed1a6cf586aa30d/tutorials_python.zip and b/master/_downloads/763344228ae6bc253ed1a6cf586aa30d/tutorials_python.zip differ diff --git a/master/_images/sphx_glr_01-vector-add_001.png b/master/_images/sphx_glr_01-vector-add_001.png index 1f6c3ce5e..38343430d 100644 Binary files a/master/_images/sphx_glr_01-vector-add_001.png and b/master/_images/sphx_glr_01-vector-add_001.png differ diff --git a/master/_images/sphx_glr_01-vector-add_thumb.png b/master/_images/sphx_glr_01-vector-add_thumb.png index 184787c44..46f161b20 100644 Binary files a/master/_images/sphx_glr_01-vector-add_thumb.png and b/master/_images/sphx_glr_01-vector-add_thumb.png differ diff --git a/master/_images/sphx_glr_02-fused-softmax_001.png b/master/_images/sphx_glr_02-fused-softmax_001.png index 7148128ff..3d6ef7ae2 100644 Binary files a/master/_images/sphx_glr_02-fused-softmax_001.png and b/master/_images/sphx_glr_02-fused-softmax_001.png differ diff --git a/master/_images/sphx_glr_02-fused-softmax_thumb.png b/master/_images/sphx_glr_02-fused-softmax_thumb.png index 22ccdb470..3e9f9170e 100644 Binary files a/master/_images/sphx_glr_02-fused-softmax_thumb.png and b/master/_images/sphx_glr_02-fused-softmax_thumb.png differ diff --git a/master/_images/sphx_glr_03-matrix-multiplication_001.png b/master/_images/sphx_glr_03-matrix-multiplication_001.png index e732f4dc4..d989c4e45 100644 Binary files a/master/_images/sphx_glr_03-matrix-multiplication_001.png and b/master/_images/sphx_glr_03-matrix-multiplication_001.png differ diff --git a/master/_images/sphx_glr_03-matrix-multiplication_thumb.png b/master/_images/sphx_glr_03-matrix-multiplication_thumb.png index 3233744b7..d14b6023d 100644 Binary files a/master/_images/sphx_glr_03-matrix-multiplication_thumb.png and b/master/_images/sphx_glr_03-matrix-multiplication_thumb.png differ diff --git a/master/_images/sphx_glr_05-layer-norm_001.png b/master/_images/sphx_glr_05-layer-norm_001.png index 8f14dd7b1..d72fdacdc 100644 Binary files a/master/_images/sphx_glr_05-layer-norm_001.png and b/master/_images/sphx_glr_05-layer-norm_001.png differ diff --git a/master/_images/sphx_glr_05-layer-norm_thumb.png b/master/_images/sphx_glr_05-layer-norm_thumb.png index a5a1b0d2c..f8546e407 100644 Binary files a/master/_images/sphx_glr_05-layer-norm_thumb.png and b/master/_images/sphx_glr_05-layer-norm_thumb.png differ diff --git a/master/_sources/getting-started/tutorials/01-vector-add.rst.txt b/master/_sources/getting-started/tutorials/01-vector-add.rst.txt index 680089380..5ddde3123 100644 --- a/master/_sources/getting-started/tutorials/01-vector-add.rst.txt +++ b/master/_sources/getting-started/tutorials/01-vector-add.rst.txt @@ -245,7 +245,7 @@ We can now run the decorated function above. Pass `print_data=True` to see the p 10 4194304.0 780.190482 780.190482 11 8388608.0 812.429770 812.429770 12 16777216.0 833.084721 833.084721 - 13 33554432.0 842.004273 842.004273 + 13 33554432.0 842.004273 843.811163 14 67108864.0 847.448255 848.362445 15 134217728.0 849.737435 850.656574 @@ -255,7 +255,7 @@ We can now run the decorated function above. Pass `print_data=True` to see the p .. rst-class:: sphx-glr-timing - **Total running time of the script:** ( 1 minutes 40.796 seconds) + **Total running time of the script:** ( 1 minutes 47.140 seconds) .. _sphx_glr_download_getting-started_tutorials_01-vector-add.py: diff --git a/master/_sources/getting-started/tutorials/02-fused-softmax.rst.txt b/master/_sources/getting-started/tutorials/02-fused-softmax.rst.txt index 01a38a639..b7b7fc8f4 100644 --- a/master/_sources/getting-started/tutorials/02-fused-softmax.rst.txt +++ b/master/_sources/getting-started/tutorials/02-fused-softmax.rst.txt @@ -278,17 +278,17 @@ We will then compare its performance against (1) :code:`torch.softmax` and (2) t softmax-performance: N Triton Torch (native) Torch (jit) - 0 256.0 512.000001 512.000001 186.181817 + 0 256.0 546.133347 512.000001 186.181817 1 384.0 585.142862 585.142862 151.703707 2 512.0 655.360017 606.814814 154.566038 - 3 640.0 682.666684 640.000002 160.000000 + 3 640.0 682.666684 640.000002 158.759699 4 768.0 722.823517 664.216187 162.754967 .. ... ... ... ... - 93 12160.0 814.058574 405.755985 198.834951 - 94 12288.0 814.111783 415.222812 199.096718 - 95 12416.0 814.163950 411.722274 198.854847 - 96 12544.0 814.214963 412.546756 198.913776 - 97 12672.0 814.265046 411.679167 199.069228 + 93 12160.0 814.058574 405.755985 198.936606 + 94 12288.0 814.111783 415.661740 198.995960 + 95 12416.0 814.163950 411.296057 198.805107 + 96 12544.0 814.214963 412.971190 198.913776 + 97 12672.0 814.265046 411.888249 198.971549 [98 rows x 4 columns] @@ -306,7 +306,7 @@ In the above plot, we can see that: .. rst-class:: sphx-glr-timing - **Total running time of the script:** ( 3 minutes 19.085 seconds) + **Total running time of the script:** ( 3 minutes 27.130 seconds) .. _sphx_glr_download_getting-started_tutorials_02-fused-softmax.py: diff --git a/master/_sources/getting-started/tutorials/03-matrix-multiplication.rst.txt b/master/_sources/getting-started/tutorials/03-matrix-multiplication.rst.txt index 7af0fab56..83692fe09 100644 --- a/master/_sources/getting-started/tutorials/03-matrix-multiplication.rst.txt +++ b/master/_sources/getting-started/tutorials/03-matrix-multiplication.rst.txt @@ -457,38 +457,38 @@ We can now compare the performance of our kernel against that of cuBLAS. Here we .. code-block:: none matmul-performance: - M cuBLAS ... Triton Triton (+ LeakyReLU) - 0 256.0 2.730667 ... 3.276800 2.978909 - 1 384.0 7.372800 ... 8.507077 8.507077 - 2 512.0 14.563555 ... 15.420235 15.420235 - 3 640.0 22.260869 ... 24.380953 24.380953 - 4 768.0 32.768000 ... 34.028308 34.028308 - 5 896.0 37.971025 ... 41.321411 40.140799 - 6 1024.0 49.932191 ... 53.773130 52.428801 - 7 1152.0 45.242181 ... 48.161033 47.396572 - 8 1280.0 51.200001 ... 57.690139 57.690139 - 9 1408.0 64.138541 ... 68.147202 68.147202 - 10 1536.0 80.430545 ... 80.430545 80.430545 - 11 1664.0 62.929456 ... 63.372618 62.929456 - 12 1792.0 72.983276 ... 63.499573 63.142831 - 13 1920.0 68.776119 ... 70.892307 70.892307 - 14 2048.0 73.584279 ... 78.398206 78.033565 - 15 2176.0 83.155572 ... 87.115360 86.739860 - 16 2304.0 68.251065 ... 77.810656 77.558029 - 17 2432.0 71.305746 ... 75.726318 75.522751 - 18 2560.0 77.833728 ... 81.715711 81.920002 - 19 2688.0 83.737433 ... 90.966561 90.316801 - 20 2816.0 84.035084 ... 83.552120 83.873477 - 21 2944.0 82.509987 ... 83.899046 84.040530 - 22 3072.0 81.589488 ... 86.184329 89.735509 - 23 3200.0 84.993363 ... 96.240602 96.240602 - 24 3328.0 83.226931 ... 85.806075 84.895397 - 25 3456.0 82.604067 ... 84.068369 88.692595 - 26 3584.0 87.211821 ... 100.017124 100.351999 - 27 3712.0 85.091436 ... 89.513749 87.094458 - 28 3840.0 84.228485 ... 92.390975 85.531326 - 29 3968.0 91.954739 ... 85.034103 91.540836 - 30 4096.0 87.495257 ... 89.597949 93.336389 + M cuBLAS ... Triton Triton (+ LeakyReLU) + 0 256.0 2.730667 ... 3.276800 2.978909 + 1 384.0 7.372800 ... 8.507077 8.507077 + 2 512.0 14.563555 ... 16.384000 16.384000 + 3 640.0 22.260869 ... 24.380953 24.380953 + 4 768.0 32.768000 ... 35.389441 34.028308 + 5 896.0 37.971025 ... 40.140799 40.140799 + 6 1024.0 49.932191 ... 53.773130 53.773130 + 7 1152.0 45.242181 ... 48.161033 47.396572 + 8 1280.0 51.200001 ... 57.690139 57.690139 + 9 1408.0 64.138541 ... 68.147202 68.147202 + 10 1536.0 80.430545 ... 81.355034 79.526831 + 11 1664.0 62.929456 ... 63.372618 62.929456 + 12 1792.0 72.983276 ... 63.499573 63.142831 + 13 1920.0 68.776119 ... 71.257735 71.257735 + 14 2048.0 73.262953 ... 78.033565 77.672296 + 15 2176.0 83.155572 ... 87.115360 86.739860 + 16 2304.0 68.251065 ... 78.064941 77.810656 + 17 2432.0 71.305746 ... 75.726318 75.522751 + 18 2560.0 77.833728 ... 82.125311 81.715711 + 19 2688.0 83.369354 ... 90.748936 90.748936 + 20 2816.0 82.290955 ... 84.523664 84.360174 + 21 2944.0 82.373605 ... 83.617504 82.373605 + 22 3072.0 82.062468 ... 89.735509 88.473602 + 23 3200.0 80.503145 ... 94.674553 95.380032 + 24 3328.0 82.369902 ... 86.320498 86.736504 + 25 3456.0 78.578525 ... 84.332184 87.823058 + 26 3584.0 87.381330 ... 99.684470 99.463928 + 27 3712.0 83.247783 ... 89.916604 87.018592 + 28 3840.0 84.874902 ... 92.275341 86.332554 + 29 3968.0 92.864488 ... 86.664727 92.512459 + 30 4096.0 86.592080 ... 87.552332 93.271527 [31 rows x 5 columns] @@ -498,7 +498,7 @@ We can now compare the performance of our kernel against that of cuBLAS. Here we .. rst-class:: sphx-glr-timing - **Total running time of the script:** ( 6 minutes 7.559 seconds) + **Total running time of the script:** ( 6 minutes 16.209 seconds) .. _sphx_glr_download_getting-started_tutorials_03-matrix-multiplication.py: diff --git a/master/_sources/getting-started/tutorials/05-layer-norm.rst.txt b/master/_sources/getting-started/tutorials/05-layer-norm.rst.txt index 98ad217ff..661a15717 100644 --- a/master/_sources/getting-started/tutorials/05-layer-norm.rst.txt +++ b/master/_sources/getting-started/tutorials/05-layer-norm.rst.txt @@ -38,36 +38,36 @@ Layer Normalization layer-norm-backward: N Triton Torch Apex - 0 1024.0 361.411758 99.902435 315.076934 + 0 1024.0 361.411758 99.497980 311.088617 1 1536.0 405.098894 133.083026 344.523365 - 2 2048.0 496.484863 158.554837 334.367350 + 2 2048.0 496.484863 158.045011 334.367350 3 2560.0 461.954908 182.857144 330.322572 4 3072.0 519.211251 191.501303 320.556515 - 5 3584.0 554.941930 208.271186 309.410081 + 5 3584.0 554.941930 208.271186 308.301075 6 4096.0 564.965515 220.412561 298.796351 - 7 4608.0 502.690905 231.849059 286.507772 - 8 5120.0 527.381977 244.294240 286.433562 - 9 5632.0 542.843364 244.426754 292.571431 - 10 6144.0 550.208948 251.202731 288.000001 - 11 6656.0 534.260858 256.000009 286.793541 - 12 7168.0 513.528374 253.734520 277.470965 - 13 7680.0 488.912481 266.743841 284.884090 - 14 8192.0 468.114289 258.694729 278.087683 - 15 8704.0 415.300208 267.472468 286.158893 - 16 9216.0 429.483477 273.066667 289.129410 - 17 9728.0 438.857162 279.942444 289.308559 - 18 10240.0 444.412281 287.102804 290.153487 - 19 10752.0 427.940303 246.935876 290.104546 - 20 11264.0 429.104745 246.432094 287.897767 - 21 11776.0 421.198220 249.888595 289.573776 + 7 4608.0 500.416301 232.336141 286.507772 + 8 5120.0 529.655159 243.809526 286.433562 + 9 5632.0 542.843364 244.426754 291.310338 + 10 6144.0 548.163546 251.202731 286.879370 + 11 6656.0 534.260858 256.410903 286.793541 + 12 7168.0 513.528374 253.360829 277.470965 + 13 7680.0 486.332448 266.743841 284.884090 + 14 8192.0 468.114289 258.694729 277.694924 + 15 8704.0 414.476194 267.472468 286.158893 + 16 9216.0 428.651187 273.066667 289.507855 + 17 9728.0 438.857162 279.942444 288.593329 + 18 10240.0 445.217381 287.102804 290.153487 + 19 10752.0 427.231788 246.699797 289.941565 + 20 11264.0 427.071098 246.432094 288.204696 + 21 11776.0 421.826879 249.667843 289.573776 22 12288.0 417.131525 254.673582 294.323369 - 23 12800.0 411.244989 253.884294 290.084977 - 24 13312.0 409.599999 252.959629 289.391298 - 25 13824.0 405.594132 256.991469 292.056329 + 23 12800.0 411.244989 253.989249 290.084977 + 24 13312.0 409.599999 253.160074 289.391298 + 25 13824.0 405.842204 256.991469 292.313649 26 14336.0 395.475867 255.619613 288.402346 - 27 14848.0 385.245405 256.922861 288.544136 - 28 15360.0 380.041240 258.332158 288.450715 - 29 15872.0 372.000001 263.071829 290.341468 + 27 14848.0 383.999990 256.922861 288.544136 + 28 15360.0 380.041240 258.513318 288.225185 + 29 15872.0 372.000001 261.446802 289.239176 @@ -339,7 +339,7 @@ Layer Normalization .. rst-class:: sphx-glr-timing - **Total running time of the script:** ( 2 minutes 13.781 seconds) + **Total running time of the script:** ( 2 minutes 13.538 seconds) .. _sphx_glr_download_getting-started_tutorials_05-layer-norm.py: diff --git a/master/_sources/getting-started/tutorials/sg_execution_times.rst.txt b/master/_sources/getting-started/tutorials/sg_execution_times.rst.txt index 895c68fba..ed0d5b820 100644 --- a/master/_sources/getting-started/tutorials/sg_execution_times.rst.txt +++ b/master/_sources/getting-started/tutorials/sg_execution_times.rst.txt @@ -5,16 +5,16 @@ Computation times ================= -**13:21.234** total execution time for **getting-started_tutorials** files: +**13:44.030** total execution time for **getting-started_tutorials** files: +---------------------------------------------------------------------------------------------------------+-----------+--------+ -| :ref:`sphx_glr_getting-started_tutorials_03-matrix-multiplication.py` (``03-matrix-multiplication.py``) | 06:07.559 | 0.0 MB | +| :ref:`sphx_glr_getting-started_tutorials_03-matrix-multiplication.py` (``03-matrix-multiplication.py``) | 06:16.209 | 0.0 MB | +---------------------------------------------------------------------------------------------------------+-----------+--------+ -| :ref:`sphx_glr_getting-started_tutorials_02-fused-softmax.py` (``02-fused-softmax.py``) | 03:19.085 | 0.0 MB | +| :ref:`sphx_glr_getting-started_tutorials_02-fused-softmax.py` (``02-fused-softmax.py``) | 03:27.130 | 0.0 MB | +---------------------------------------------------------------------------------------------------------+-----------+--------+ -| :ref:`sphx_glr_getting-started_tutorials_05-layer-norm.py` (``05-layer-norm.py``) | 02:13.781 | 0.0 MB | +| :ref:`sphx_glr_getting-started_tutorials_05-layer-norm.py` (``05-layer-norm.py``) | 02:13.538 | 0.0 MB | +---------------------------------------------------------------------------------------------------------+-----------+--------+ -| :ref:`sphx_glr_getting-started_tutorials_01-vector-add.py` (``01-vector-add.py``) | 01:40.796 | 0.0 MB | +| :ref:`sphx_glr_getting-started_tutorials_01-vector-add.py` (``01-vector-add.py``) | 01:47.140 | 0.0 MB | +---------------------------------------------------------------------------------------------------------+-----------+--------+ | :ref:`sphx_glr_getting-started_tutorials_04-low-memory-dropout.py` (``04-low-memory-dropout.py``) | 00:00.012 | 0.0 MB | +---------------------------------------------------------------------------------------------------------+-----------+--------+ diff --git a/master/getting-started/tutorials/01-vector-add.html b/master/getting-started/tutorials/01-vector-add.html index f49d233b2..58398594e 100644 --- a/master/getting-started/tutorials/01-vector-add.html +++ b/master/getting-started/tutorials/01-vector-add.html @@ -335,12 +335,12 @@ for different problem sizes.

10 4194304.0 780.190482 780.190482 11 8388608.0 812.429770 812.429770 12 16777216.0 833.084721 833.084721 -13 33554432.0 842.004273 842.004273 +13 33554432.0 842.004273 843.811163 14 67108864.0 847.448255 848.362445 15 134217728.0 849.737435 850.656574 -

Total running time of the script: ( 1 minutes 40.796 seconds)

+

Total running time of the script: ( 1 minutes 47.140 seconds)