diff --git a/_downloads/662999063954282841dc90b8945f85ce/tutorials_jupyter.zip b/_downloads/662999063954282841dc90b8945f85ce/tutorials_jupyter.zip index 82e42c143..aab318caa 100644 Binary files a/_downloads/662999063954282841dc90b8945f85ce/tutorials_jupyter.zip and b/_downloads/662999063954282841dc90b8945f85ce/tutorials_jupyter.zip differ diff --git a/_downloads/763344228ae6bc253ed1a6cf586aa30d/tutorials_python.zip b/_downloads/763344228ae6bc253ed1a6cf586aa30d/tutorials_python.zip index f3dc2e234..a4dd8413e 100644 Binary files a/_downloads/763344228ae6bc253ed1a6cf586aa30d/tutorials_python.zip and b/_downloads/763344228ae6bc253ed1a6cf586aa30d/tutorials_python.zip differ diff --git a/_images/sphx_glr_01-vector-add_001.png b/_images/sphx_glr_01-vector-add_001.png index 67b83135a..5723bca13 100644 Binary files a/_images/sphx_glr_01-vector-add_001.png and b/_images/sphx_glr_01-vector-add_001.png differ diff --git a/_images/sphx_glr_01-vector-add_thumb.png b/_images/sphx_glr_01-vector-add_thumb.png index 318ba9d77..96771572d 100644 Binary files a/_images/sphx_glr_01-vector-add_thumb.png and b/_images/sphx_glr_01-vector-add_thumb.png differ diff --git a/_images/sphx_glr_02-fused-softmax_001.png b/_images/sphx_glr_02-fused-softmax_001.png index 6f4033674..5f3cc3772 100644 Binary files a/_images/sphx_glr_02-fused-softmax_001.png and b/_images/sphx_glr_02-fused-softmax_001.png differ diff --git a/_images/sphx_glr_02-fused-softmax_thumb.png b/_images/sphx_glr_02-fused-softmax_thumb.png index deaa8e1d8..35ee4dd0a 100644 Binary files a/_images/sphx_glr_02-fused-softmax_thumb.png and b/_images/sphx_glr_02-fused-softmax_thumb.png differ diff --git a/_images/sphx_glr_03-matrix-multiplication_001.png b/_images/sphx_glr_03-matrix-multiplication_001.png index 04c7e5d6d..e8bee9f60 100644 Binary files a/_images/sphx_glr_03-matrix-multiplication_001.png and b/_images/sphx_glr_03-matrix-multiplication_001.png differ diff --git a/_images/sphx_glr_03-matrix-multiplication_thumb.png b/_images/sphx_glr_03-matrix-multiplication_thumb.png index 3eea650d9..b1e6e6da9 100644 Binary files a/_images/sphx_glr_03-matrix-multiplication_thumb.png and b/_images/sphx_glr_03-matrix-multiplication_thumb.png differ diff --git a/_sources/getting-started/tutorials/01-vector-add.rst.txt b/_sources/getting-started/tutorials/01-vector-add.rst.txt index bdfbd6a6f..a165f60b2 100644 --- a/_sources/getting-started/tutorials/01-vector-add.rst.txt +++ b/_sources/getting-started/tutorials/01-vector-add.rst.txt @@ -233,8 +233,8 @@ We can now run the decorated function above. Pass `print_data=True` to see the p size Triton Torch 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 + 2 16384.0 38.400001 31.999999 + 3 32768.0 63.999998 76.800002 4 65536.0 127.999995 127.999995 5 131072.0 219.428568 219.428568 6 262144.0 341.333321 384.000001 @@ -244,7 +244,7 @@ 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 @@ -254,7 +254,7 @@ 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 53.211 seconds) + **Total running time of the script:** ( 1 minutes 52.287 seconds) .. _sphx_glr_download_getting-started_tutorials_01-vector-add.py: diff --git a/_sources/getting-started/tutorials/02-fused-softmax.rst.txt b/_sources/getting-started/tutorials/02-fused-softmax.rst.txt index 9e1e07d3e..4c7b124f7 100644 --- a/_sources/getting-started/tutorials/02-fused-softmax.rst.txt +++ b/_sources/getting-started/tutorials/02-fused-softmax.rst.txt @@ -286,17 +286,17 @@ 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 190.511628 - 1 384.0 585.142862 585.142862 151.703707 - 2 512.0 655.360017 606.814814 156.038096 + 0 256.0 512.000001 546.133347 186.181817 + 1 384.0 585.142862 585.142862 153.600004 + 2 512.0 630.153853 606.814814 154.566038 3 640.0 682.666684 640.000002 160.000000 4 768.0 702.171410 664.216187 163.839992 .. ... ... ... ... - 93 12160.0 810.666687 405.755985 198.936606 - 94 12288.0 810.754644 415.661740 199.298541 - 95 12416.0 809.189387 411.722274 198.904612 + 93 12160.0 810.666687 406.179533 199.140227 + 94 12288.0 810.754644 415.661740 199.399583 + 95 12416.0 809.189387 412.149375 198.954424 96 12544.0 807.661970 412.971190 199.209928 - 97 12672.0 807.776923 411.679167 199.264875 + 97 12672.0 807.776923 412.097543 199.264875 [98 rows x 4 columns] @@ -314,7 +314,7 @@ In the above plot, we can see that: .. rst-class:: sphx-glr-timing - **Total running time of the script:** ( 3 minutes 31.670 seconds) + **Total running time of the script:** ( 3 minutes 29.511 seconds) .. _sphx_glr_download_getting-started_tutorials_02-fused-softmax.py: diff --git a/_sources/getting-started/tutorials/03-matrix-multiplication.rst.txt b/_sources/getting-started/tutorials/03-matrix-multiplication.rst.txt index 5b08d800d..9acb6c6a3 100644 --- a/_sources/getting-started/tutorials/03-matrix-multiplication.rst.txt +++ b/_sources/getting-started/tutorials/03-matrix-multiplication.rst.txt @@ -463,36 +463,36 @@ We can now compare the performance of our kernel against that of cuBLAS. Here we matmul-performance: M cuBLAS ... Triton Triton (+ LeakyReLU) 0 256.0 2.730667 ... 2.978909 2.978909 - 1 384.0 7.372800 ... 8.507077 8.507077 + 1 384.0 7.372800 ... 8.507077 7.899428 2 512.0 14.563555 ... 16.384000 16.384000 3 640.0 22.260869 ... 24.380953 24.380953 4 768.0 32.768000 ... 34.028308 34.028308 - 5 896.0 39.025776 ... 39.025776 39.025776 - 6 1024.0 49.932191 ... 53.773130 52.428801 - 7 1152.0 44.566925 ... 46.656000 46.656000 + 5 896.0 39.025776 ... 39.025776 37.971025 + 6 1024.0 49.932191 ... 52.428801 52.428801 + 7 1152.0 44.566925 ... 46.656000 45.938215 8 1280.0 51.200001 ... 56.109587 56.109587 9 1408.0 64.138541 ... 66.485074 66.485074 - 10 1536.0 80.430545 ... 78.643199 78.643199 - 11 1664.0 62.929456 ... 62.061463 62.061463 - 12 1792.0 72.983276 ... 72.047592 72.047592 - 13 1920.0 69.120002 ... 70.172588 70.172588 - 14 2048.0 73.908442 ... 76.608294 76.260072 - 15 2176.0 82.813365 ... 85.998493 85.269692 - 16 2304.0 68.446623 ... 76.563695 76.563695 - 17 2432.0 71.305746 ... 74.521127 85.393507 - 18 2560.0 77.833728 ... 81.310171 81.310171 - 19 2688.0 83.552988 ... 89.464755 89.044730 - 20 2816.0 83.712490 ... 83.552120 82.916747 - 21 2944.0 82.373605 ... 83.060049 82.102191 - 22 3072.0 81.121923 ... 89.170242 88.197981 - 23 3200.0 84.880639 ... 95.238096 94.534716 - 24 3328.0 83.130825 ... 83.905938 84.200347 - 25 3456.0 82.015834 ... 90.994998 85.585527 - 26 3584.0 85.797134 ... 91.750399 94.947616 - 27 3712.0 80.823095 ... 87.475786 88.015279 - 28 3840.0 83.718392 ... 91.097196 86.602979 - 29 3968.0 87.818595 ... 90.994735 86.973584 - 30 4096.0 93.142072 ... 88.417474 85.325956 + 10 1536.0 79.526831 ... 79.526831 78.643199 + 11 1664.0 62.929456 ... 62.492442 62.061463 + 12 1792.0 72.983276 ... 72.512412 72.047592 + 13 1920.0 69.120002 ... 70.530615 70.172588 + 14 2048.0 73.908442 ... 76.959706 76.608294 + 15 2176.0 83.500614 ... 85.632545 85.269692 + 16 2304.0 68.446623 ... 76.809875 76.809875 + 17 2432.0 71.305746 ... 84.621881 84.877538 + 18 2560.0 77.649287 ... 80.709358 81.310171 + 19 2688.0 83.552988 ... 89.464755 89.464755 + 20 2816.0 83.392363 ... 83.392363 82.916747 + 21 2944.0 82.509987 ... 82.373605 82.237674 + 22 3072.0 82.301023 ... 88.750943 88.335577 + 23 3200.0 84.993363 ... 95.380032 94.955488 + 24 3328.0 83.130825 ... 84.200347 84.895397 + 25 3456.0 81.108217 ... 85.858966 85.494768 + 26 3584.0 86.707226 ... 97.947050 98.429551 + 27 3712.0 81.548851 ... 88.015279 88.326564 + 28 3840.0 80.139129 ... 89.187096 91.097196 + 29 3968.0 85.992909 ... 87.347124 88.103928 + 30 4096.0 91.805174 ... 92.948562 85.216761 [31 rows x 5 columns] @@ -502,7 +502,7 @@ 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 36.541 seconds) + **Total running time of the script:** ( 6 minutes 28.511 seconds) .. _sphx_glr_download_getting-started_tutorials_03-matrix-multiplication.py: diff --git a/_sources/getting-started/tutorials/04-low-memory-dropout.rst.txt b/_sources/getting-started/tutorials/04-low-memory-dropout.rst.txt index 34799280f..0ae825e4e 100644 --- a/_sources/getting-started/tutorials/04-low-memory-dropout.rst.txt +++ b/_sources/getting-started/tutorials/04-low-memory-dropout.rst.txt @@ -238,7 +238,7 @@ References .. rst-class:: sphx-glr-timing - **Total running time of the script:** ( 0 minutes 0.195 seconds) + **Total running time of the script:** ( 0 minutes 0.354 seconds) .. _sphx_glr_download_getting-started_tutorials_04-low-memory-dropout.py: diff --git a/_sources/getting-started/tutorials/sg_execution_times.rst.txt b/_sources/getting-started/tutorials/sg_execution_times.rst.txt index 9c0dcc5c8..02e8216a4 100644 --- a/_sources/getting-started/tutorials/sg_execution_times.rst.txt +++ b/_sources/getting-started/tutorials/sg_execution_times.rst.txt @@ -5,14 +5,14 @@ Computation times ================= -**12:01.617** total execution time for **getting-started_tutorials** files: +**11:50.662** total execution time for **getting-started_tutorials** files: +---------------------------------------------------------------------------------------------------------+-----------+--------+ -| :ref:`sphx_glr_getting-started_tutorials_03-matrix-multiplication.py` (``03-matrix-multiplication.py``) | 06:36.541 | 0.0 MB | +| :ref:`sphx_glr_getting-started_tutorials_03-matrix-multiplication.py` (``03-matrix-multiplication.py``) | 06:28.511 | 0.0 MB | +---------------------------------------------------------------------------------------------------------+-----------+--------+ -| :ref:`sphx_glr_getting-started_tutorials_02-fused-softmax.py` (``02-fused-softmax.py``) | 03:31.670 | 0.0 MB | +| :ref:`sphx_glr_getting-started_tutorials_02-fused-softmax.py` (``02-fused-softmax.py``) | 03:29.511 | 0.0 MB | +---------------------------------------------------------------------------------------------------------+-----------+--------+ -| :ref:`sphx_glr_getting-started_tutorials_01-vector-add.py` (``01-vector-add.py``) | 01:53.211 | 0.0 MB | +| :ref:`sphx_glr_getting-started_tutorials_01-vector-add.py` (``01-vector-add.py``) | 01:52.287 | 0.0 MB | +---------------------------------------------------------------------------------------------------------+-----------+--------+ -| :ref:`sphx_glr_getting-started_tutorials_04-low-memory-dropout.py` (``04-low-memory-dropout.py``) | 00:00.195 | 0.0 MB | +| :ref:`sphx_glr_getting-started_tutorials_04-low-memory-dropout.py` (``04-low-memory-dropout.py``) | 00:00.354 | 0.0 MB | +---------------------------------------------------------------------------------------------------------+-----------+--------+ diff --git a/getting-started/tutorials/01-vector-add.html b/getting-started/tutorials/01-vector-add.html index 70cd3bbc9..143b47e5f 100644 --- a/getting-started/tutorials/01-vector-add.html +++ b/getting-started/tutorials/01-vector-add.html @@ -322,8 +322,8 @@ for different problem sizes.
size Triton Torch 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 +2 16384.0 38.400001 31.999999 +3 32768.0 63.999998 76.800002 4 65536.0 127.999995 127.999995 5 131072.0 219.428568 219.428568 6 262144.0 341.333321 384.000001 @@ -333,12 +333,12 @@ 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 53.211 seconds)
+Total running time of the script: ( 1 minutes 52.287 seconds)