diff --git a/master/.buildinfo b/master/.buildinfo index fb7c524af..08dd12f3b 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: 129fe056d957bceb5eb7005a87aa6582 +config: 5d1e5336b0fe1254665c1a22d9c46408 tags: 645f666f9bcd5a90fca523b33c5a78b7 diff --git a/master/.doctrees/environment.pickle b/master/.doctrees/environment.pickle index 22ce875ae..94eb69708 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 69f41b558..d77309df7 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 9c6eadb12..537eaeae3 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 f45a0662d..ea9a22638 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 4a7f92923..5f86ffa16 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 5c33be468..a0d44397b 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 5b612456d..ef52d63d1 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 2eb7905fa..737d02547 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 c04fcff95..e6da6530f 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 1ba38d7dd..3a229ff1c 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 4a886c025..dc97f532e 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 794bb44cc..e5e15d601 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 52bf7cc02..db8e4e9ff 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 0537a7ebd..37a2da6a6 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 e2f4ba1e6..34d6a474a 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 1a3dac78a..a4a74df82 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 4337f62ff..59b2bb7fc 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 b223291ff..4a026e3eb 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 5999317c0..e403e2005 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 c5106911e..538a31622 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 9d298744f..23badf315 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 157f636cd..10167f7d2 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 94a11a263..50fe71f81 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 b7bef27bd..f25abfe2c 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 da0d5dafd..d0c862025 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 73452a9b1..6eb4519e4 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 c339fded6..d05a98e7b 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 376937376..3975fb810 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 dfb589ed6..445862c9b 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 9884d7611..bb9203f5a 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 51b559a84..d6d0284db 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 853ebc899..e163ccc4b 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 fd15fdd77..e19b565b5 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 014fccdf9..8a435cd66 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 27d26fa16..a3f2fa289 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 7056cffff..7f7444ffe 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 1d409a3d6..291b4f0f5 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 20f5fbc6d..5cdfabfcf 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 a7d702e74..0db2d6093 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 b124d3a4f..7db8a3e5e 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 210c3b64a..39781c95a 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 8f55d7fb9..ee39c71fe 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 952ccff31..aa8f7a4fa 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 931e58709..af796393b 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 7b143a344..34084dc96 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 9a8088019..b362a283a 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 deda59db3..22d4a2abf 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 eb90ee6ab..3514e2e2e 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 dfa29de29..8e7de10a4 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 493a0badd..a5ad0abee 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 b95b05649..b5b4716b9 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 876e09aa2..6a1777038 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 fbb48fd5d..06a01cb1f 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 1665f687c..6b7737460 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 3e8e256ce..945fbf47e 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 7dec37ece..c8cb5564e 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 93b0c5a77..5517fab6b 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 dcf198269..5a990ce42 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 b02d989c3..b04c489ee 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 ff114e19b..de340db21 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 3b1becdc4..db485525f 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 f1ba6c1a7..44463ebc3 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 02ff4e548..4ee6a0715 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 6ce0b41cc..f07cee5b7 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 787da668f..9737bca4d 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 fe069bd79..6398d5a48 100644 --- a/master/_sources/getting-started/tutorials/01-vector-add.rst.txt +++ b/master/_sources/getting-started/tutorials/01-vector-add.rst.txt @@ -232,16 +232,16 @@ We can now run the decorated function above. Pass `print_data=True` to see the p vector-add-performance: size Triton Torch - 0 4096.0 8.000000 9.600000 + 0 4096.0 9.600000 9.600000 1 8192.0 19.200000 19.200000 2 16384.0 38.400001 38.400001 - 3 32768.0 76.800002 76.800002 + 3 32768.0 63.999998 63.999998 4 65536.0 127.999995 127.999995 5 131072.0 219.428568 219.428568 6 262144.0 341.333321 384.000001 7 524288.0 472.615390 472.615390 8 1048576.0 614.400016 614.400016 - 9 2097152.0 722.823517 722.823517 + 9 2097152.0 722.823517 702.171410 10 4194304.0 780.190482 780.190482 11 8388608.0 812.429770 812.429770 12 16777216.0 833.084721 833.084721 @@ -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 46.935 seconds) + **Total running time of the script:** ( 1 minutes 24.953 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 54858bd9f..c05d61ac5 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 546.133347 188.321838 + 0 256.0 512.000001 546.133347 190.511628 1 384.0 585.142862 558.545450 151.703707 2 512.0 655.360017 606.814814 154.566038 3 640.0 682.666684 640.000002 158.759699 4 768.0 722.823517 664.216187 163.839992 .. ... ... ... ... - 93 12160.0 814.058574 405.755985 198.733401 - 94 12288.0 814.111783 415.661740 198.995960 - 95 12416.0 814.163950 411.722274 198.705656 - 96 12544.0 814.214963 412.971190 198.815254 - 97 12672.0 814.265046 411.679167 198.971549 + 93 12160.0 814.058574 405.755985 198.530610 + 94 12288.0 814.111783 415.222812 198.794749 + 95 12416.0 814.163950 412.149375 198.457532 + 96 12544.0 814.214963 412.971190 198.618504 + 97 12672.0 814.265046 411.679167 198.776477 [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 27.513 seconds) + **Total running time of the script:** ( 3 minutes 19.032 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 f3a2dc1a2..47fc37fd4 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 ... 2.978909 2.978909 - 1 384.0 7.372800 ... 7.899428 7.899428 - 2 512.0 14.563555 ... 15.420235 15.420235 - 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 39.025776 - 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 ... 69.009825 68.147202 - 10 1536.0 80.430545 ... 80.430545 80.430545 - 11 1664.0 63.372618 ... 63.372618 63.372618 - 12 1792.0 72.983276 ... 63.499573 63.142831 - 13 1920.0 69.120002 ... 71.626943 70.892307 - 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 77.558029 - 17 2432.0 71.305746 ... 75.726318 75.522751 - 18 2560.0 77.833728 ... 82.125311 82.125311 - 19 2688.0 83.552988 ... 90.316801 90.966561 - 20 2816.0 83.233226 ... 83.233226 84.278666 - 21 2944.0 81.166173 ... 83.617504 84.182483 - 22 3072.0 81.825298 ... 89.451983 89.451983 - 23 3200.0 83.989503 ... 96.530922 94.256261 - 24 3328.0 83.905938 ... 86.946008 86.217120 - 25 3456.0 80.300370 ... 92.350019 88.497878 - 26 3584.0 87.381330 ... 98.053863 99.463928 - 27 3712.0 85.528545 ... 87.629253 88.326564 - 28 3840.0 82.716526 ... 88.971840 91.625518 - 29 3968.0 87.004591 ... 92.652949 85.841672 - 30 4096.0 93.727466 ... 89.092421 87.495257 + M cuBLAS ... Triton Triton (+ LeakyReLU) + 0 256.0 2.730667 ... 2.978909 3.276800 + 1 384.0 7.372800 ... 7.899428 7.899428 + 2 512.0 14.563555 ... 15.420235 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 ... 69.009825 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 ... 71.257735 71.257735 + 14 2048.0 73.584279 ... 78.398206 78.033565 + 15 2176.0 83.500614 ... 87.115360 86.739860 + 16 2304.0 68.251065 ... 78.320893 77.810656 + 17 2432.0 71.305746 ... 75.726318 75.522751 + 18 2560.0 77.833728 ... 82.331658 81.715711 + 19 2688.0 83.552988 ... 90.966561 90.966561 + 20 2816.0 82.602666 ... 84.687779 84.523664 + 21 2944.0 82.921853 ... 83.758038 84.182483 + 22 3072.0 82.181572 ... 89.170242 88.197981 + 23 3200.0 81.321474 ... 95.238096 94.395283 + 24 3328.0 82.939284 ... 86.320498 85.908470 + 25 3456.0 80.140726 ... 87.727494 91.407671 + 26 3584.0 87.381330 ... 100.128496 99.684470 + 27 3712.0 83.247783 ... 87.629253 89.194055 + 28 3840.0 84.228485 ... 92.236860 85.533390 + 29 3968.0 91.885495 ... 85.093402 90.589410 + 30 4096.0 91.616198 ... 93.792965 91.242506 [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 7.255 seconds) + **Total running time of the script:** ( 5 minutes 44.529 seconds) .. _sphx_glr_download_getting-started_tutorials_03-matrix-multiplication.py: diff --git a/master/_sources/getting-started/tutorials/04-low-memory-dropout.rst.txt b/master/_sources/getting-started/tutorials/04-low-memory-dropout.rst.txt index 1644ef41a..8c652cdd2 100644 --- a/master/_sources/getting-started/tutorials/04-low-memory-dropout.rst.txt +++ b/master/_sources/getting-started/tutorials/04-low-memory-dropout.rst.txt @@ -240,7 +240,7 @@ References .. rst-class:: sphx-glr-timing - **Total running time of the script:** ( 0 minutes 0.014 seconds) + **Total running time of the script:** ( 0 minutes 0.013 seconds) .. _sphx_glr_download_getting-started_tutorials_04-low-memory-dropout.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 e8154c5c6..c5efc1510 100644 --- a/master/_sources/getting-started/tutorials/05-layer-norm.rst.txt +++ b/master/_sources/getting-started/tutorials/05-layer-norm.rst.txt @@ -39,35 +39,35 @@ Layer Normalization layer-norm-backward: N Triton Torch Apex 0 1024.0 361.411758 97.912354 303.407414 - 1 1536.0 405.098894 134.540150 341.333333 + 1 1536.0 409.599994 134.540150 341.333333 2 2048.0 491.520012 161.154101 334.367350 - 3 2560.0 465.454542 181.238943 330.322572 + 3 2560.0 461.954908 181.238943 330.322572 4 3072.0 519.211251 192.501302 323.368415 5 3584.0 558.545477 208.271186 311.652167 - 6 4096.0 564.965515 220.907859 298.796351 + 6 4096.0 568.231237 220.907859 298.796351 7 4608.0 502.690905 232.825259 287.251954 8 5120.0 527.381977 242.366855 284.444444 9 5632.0 542.843364 243.107920 289.438969 10 6144.0 546.133354 248.661056 286.322318 11 6656.0 527.207907 256.000009 285.767438 12 7168.0 503.017523 259.867079 284.821192 - 13 7680.0 482.513091 263.314295 280.121579 - 14 8192.0 463.698115 266.767970 284.526763 + 13 7680.0 483.779539 262.938666 280.121579 + 14 8192.0 463.698115 266.587109 284.526763 15 8704.0 414.476194 267.815384 284.599455 - 16 9216.0 427.822068 271.724806 287.999990 - 17 9728.0 437.213490 280.615388 290.027323 - 18 10240.0 446.836366 286.433562 290.496460 - 19 10752.0 429.364408 246.935876 290.267711 - 20 11264.0 426.397479 245.313973 286.980888 - 21 11776.0 420.571432 249.667843 288.981596 - 22 12288.0 416.542386 254.673582 294.323369 - 23 12800.0 411.244989 253.779426 289.811310 + 16 9216.0 426.996150 271.391419 287.999990 + 17 9728.0 437.213490 280.278512 290.027323 + 18 10240.0 446.025405 286.433562 290.496460 + 19 10752.0 429.364408 247.172406 290.594591 + 20 11264.0 427.071098 245.313973 286.980888 + 21 11776.0 421.198220 249.667843 289.277383 + 22 12288.0 416.542386 254.893699 294.617366 + 23 12800.0 411.244989 253.884294 289.811310 24 13312.0 409.075539 252.959629 290.443638 25 13824.0 405.098897 257.390218 292.056329 - 26 14336.0 395.021816 255.051144 286.719986 - 27 14848.0 386.080180 257.852379 289.481735 - 28 15360.0 380.433442 257.970599 286.879376 - 29 15872.0 371.094003 261.626369 289.679087 + 26 14336.0 395.021816 254.862216 287.198654 + 27 14848.0 386.498925 257.852379 289.717061 + 28 15360.0 380.433442 257.970599 287.550706 + 29 15872.0 370.552519 261.626369 290.120338 @@ -339,7 +339,7 @@ Layer Normalization .. rst-class:: sphx-glr-timing - **Total running time of the script:** ( 2 minutes 14.074 seconds) + **Total running time of the script:** ( 2 minutes 12.505 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 4b7972935..3ae762790 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 ================= -**13:35.790** total execution time for **getting-started_tutorials** files: +**12:41.032** total execution time for **getting-started_tutorials** files: +---------------------------------------------------------------------------------------------------------+-----------+--------+ -| :ref:`sphx_glr_getting-started_tutorials_03-matrix-multiplication.py` (``03-matrix-multiplication.py``) | 06:07.255 | 0.0 MB | +| :ref:`sphx_glr_getting-started_tutorials_03-matrix-multiplication.py` (``03-matrix-multiplication.py``) | 05:44.529 | 0.0 MB | +---------------------------------------------------------------------------------------------------------+-----------+--------+ -| :ref:`sphx_glr_getting-started_tutorials_02-fused-softmax.py` (``02-fused-softmax.py``) | 03:27.513 | 0.0 MB | +| :ref:`sphx_glr_getting-started_tutorials_02-fused-softmax.py` (``02-fused-softmax.py``) | 03:19.032 | 0.0 MB | +---------------------------------------------------------------------------------------------------------+-----------+--------+ -| :ref:`sphx_glr_getting-started_tutorials_05-layer-norm.py` (``05-layer-norm.py``) | 02:14.074 | 0.0 MB | +| :ref:`sphx_glr_getting-started_tutorials_05-layer-norm.py` (``05-layer-norm.py``) | 02:12.505 | 0.0 MB | +---------------------------------------------------------------------------------------------------------+-----------+--------+ -| :ref:`sphx_glr_getting-started_tutorials_01-vector-add.py` (``01-vector-add.py``) | 01:46.935 | 0.0 MB | +| :ref:`sphx_glr_getting-started_tutorials_01-vector-add.py` (``01-vector-add.py``) | 01:24.953 | 0.0 MB | +---------------------------------------------------------------------------------------------------------+-----------+--------+ -| :ref:`sphx_glr_getting-started_tutorials_04-low-memory-dropout.py` (``04-low-memory-dropout.py``) | 00:00.014 | 0.0 MB | +| :ref:`sphx_glr_getting-started_tutorials_04-low-memory-dropout.py` (``04-low-memory-dropout.py``) | 00:00.013 | 0.0 MB | +---------------------------------------------------------------------------------------------------------+-----------+--------+ diff --git a/master/getting-started/tutorials/01-vector-add.html b/master/getting-started/tutorials/01-vector-add.html index 2f023868b..1d945092c 100644 --- a/master/getting-started/tutorials/01-vector-add.html +++ b/master/getting-started/tutorials/01-vector-add.html @@ -322,16 +322,16 @@ for different problem sizes.

Out:

vector-add-performance:
            size      Triton       Torch
-0        4096.0    8.000000    9.600000
+0        4096.0    9.600000    9.600000
 1        8192.0   19.200000   19.200000
 2       16384.0   38.400001   38.400001
-3       32768.0   76.800002   76.800002
+3       32768.0   63.999998   63.999998
 4       65536.0  127.999995  127.999995
 5      131072.0  219.428568  219.428568
 6      262144.0  341.333321  384.000001
 7      524288.0  472.615390  472.615390
 8     1048576.0  614.400016  614.400016
-9     2097152.0  722.823517  722.823517
+9     2097152.0  722.823517  702.171410
 10    4194304.0  780.190482  780.190482
 11    8388608.0  812.429770  812.429770
 12   16777216.0  833.084721  833.084721
@@ -340,7 +340,7 @@ for different problem sizes.

 15  134217728.0  849.737435  850.656574
-

Total running time of the script: ( 1 minutes 46.935 seconds)

+

Total running time of the script: ( 1 minutes 24.953 seconds)