gaspersic/triton
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

[GH-PAGES] Updated website

Browse Source
This commit is contained in:
Philippe Tillet
2022-04-16 00:44:44 +00:00
parent 824d060dfb
commit 9b0ee317d9
160 changed files with 365 additions and 380 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
v1.1.2/getting-started/tutorials/01-vector-add.html
View File

@@ -322,7 +322,7 @@ for different problem sizes.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>vector-add-performance:
size Triton Torch
0 4096.0 9.600000 9.600000
1 8192.0 19.200000 19.200000
1 8192.0 15.999999 19.200000
2 16384.0 38.400001 38.400001
3 32768.0 76.800002 76.800002
4 65536.0 127.999995 127.999995
@@ -336,10 +336,10 @@ for different problem sizes.</p>
12 16777216.0 833.084721 833.084721
13 33554432.0 842.004273 842.004273
14 67108864.0 847.448255 848.362445
15 134217728.0 850.196756 850.656574
15 134217728.0 849.737435 850.656574
</pre></div>
</div>
<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 44.439 seconds)</p>
<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 44.691 seconds)</p>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-getting-started-tutorials-01-vector-add-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/62d97d49a32414049819dd8bb8378080/01-vector-add.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">01-vector-add.py</span></code></a></p>

18
v1.1.2/getting-started/tutorials/02-fused-softmax.html
View File

@@ -374,16 +374,16 @@ We will then compare its performance against (1) <code class="code docutils lite
<p class="sphx-glr-script-out">Out:</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>softmax-performance:
N Triton Torch (native) Torch (jit)
0 256.0 512.000001 546.133347 188.321838
0 256.0 512.000001 546.133347 186.181817
1 384.0 585.142862 585.142862 151.703707
2 512.0 655.360017 585.142849 154.566038
3 640.0 682.666684 640.000002 158.759699
4 768.0 722.823517 646.736871 162.754967
2 512.0 655.360017 606.814814 156.038096
3 640.0 682.666684 640.000002 160.000000
4 768.0 722.823517 646.736871 163.839992
.. ... ... ... ...
93 12160.0 812.359066 406.179533 198.834951
94 12288.0 814.111783 416.101597 199.096718
95 12416.0 814.163950 412.577363 198.755369
96 12544.0 812.566838 413.183734 198.913776
93 12160.0 812.359066 405.755985 198.936606
94 12288.0 814.111783 415.222812 199.096718
95 12416.0 812.498981 412.149375 198.854847
96 12544.0 812.566838 412.971190 199.012395
97 12672.0 812.633240 412.097543 199.069228
[98 rows x 4 columns]
@@ -397,7 +397,7 @@ We will then compare its performance against (1) <code class="code docutils lite
Note however that the PyTorch <cite>softmax</cite> operation is more general and will works on tensors of any shape.</p></li>
</ul>
</div></blockquote>
<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 3 minutes 25.535 seconds)</p>
<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 3 minutes 24.321 seconds)</p>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-getting-started-tutorials-02-fused-softmax-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/d91442ac2982c4e0cc3ab0f43534afbc/02-fused-softmax.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">02-fused-softmax.py</span></code></a></p>

58
v1.1.2/getting-started/tutorials/03-matrix-multiplication.html
View File

@@ -569,41 +569,41 @@ torch_output=tensor([[ 1.1045, -36.9688, 31.4688, ..., -11.3906, 24.4531, -3
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>matmul-performance:
M cuBLAS ... Triton Triton (+ LeakyReLU)
0 256.0 2.730667 ... 2.978909 2.978909
1 384.0 7.372800 ... 8.507077 7.899428
2 512.0 14.563555 ... 16.384000 15.420235
3 640.0 22.260869 ... 24.380953 23.272727
4 768.0 31.597714 ... 34.028308 34.028308
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 ... 34.028308 34.028308
5 896.0 37.971025 ... 39.025776 37.971025
6 1024.0 49.932191 ... 52.428801 51.150050
7 1152.0 43.911529 ... 45.938215 45.242181
8 1280.0 49.951220 ... 55.351349 55.351349
9 1408.0 62.664092 ... 65.684049 65.684049
10 1536.0 78.643199 ... 77.778988 77.778988
11 1664.0 62.061463 ... 61.217089 60.803457
12 1792.0 71.588687 ... 71.135597 70.688200
13 1920.0 68.098521 ... 69.120002 69.120002
14 2048.0 72.628641 ... 75.573044 75.234154
15 2176.0 81.803444 ... 84.199364 83.500614
16 2304.0 67.100763 ... 75.119093 75.119093
17 2432.0 69.886725 ... 83.614477 83.119713
18 2560.0 76.204654 ... 79.727497 79.341404
19 2688.0 82.284288 ... 88.216412 88.011732
20 2816.0 82.759409 ... 82.135981 81.674548
21 2944.0 81.166173 ... 81.431424 81.166173
22 3072.0 81.121923 ... 87.381335 86.845249
23 3200.0 83.660130 ... 94.395283 93.910490
24 3328.0 82.181847 ... 83.808259 83.565058
25 3456.0 80.864158 ... 90.484366 90.079964
26 3584.0 86.291162 ... 97.628001 97.575029
27 3712.0 84.730571 ... 87.860458 87.552452
28 3840.0 84.228485 ... 91.322872 90.649182
29 3968.0 92.442373 ... 90.522206 90.354633
30 4096.0 93.142072 ... 91.929947 92.182504
7 1152.0 44.566925 ... 46.656000 46.656000
8 1280.0 51.200001 ... 56.888887 56.109587
9 1408.0 64.138541 ... 67.305878 66.485074
10 1536.0 79.526831 ... 79.526831 78.643199
11 1664.0 63.372618 ... 62.492442 62.492442
12 1792.0 72.983276 ... 72.047592 71.588687
13 1920.0 68.776119 ... 70.172588 70.172588
14 2048.0 73.908442 ... 76.959706 76.608294
15 2176.0 83.155572 ... 85.998493 85.269692
16 2304.0 68.251065 ... 76.441192 76.563695
17 2432.0 71.305746 ... 79.587714 84.621881
18 2560.0 77.833728 ... 80.908642 80.313727
19 2688.0 83.369354 ... 89.676257 89.254248
20 2816.0 84.035084 ... 82.759409 82.602666
21 2944.0 82.373605 ... 83.198715 82.921853
22 3072.0 81.943708 ... 87.855861 88.335577
23 3200.0 84.321474 ... 89.761569 94.674553
24 3328.0 82.939284 ... 79.548391 83.710812
25 3456.0 81.683457 ... 89.480098 90.790053
26 3584.0 87.211821 ... 98.591437 90.367227
27 3712.0 85.019017 ... 88.718781 84.017953
28 3840.0 84.679936 ... 92.390975 84.228485
29 3968.0 92.723355 ... 85.271796 89.657558
30 4096.0 92.787924 ... 92.372834 86.258181
[31 rows x 5 columns]
</pre></div>
</div>
<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 5 minutes 23.021 seconds)</p>
<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 5 minutes 34.991 seconds)</p>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-getting-started-tutorials-03-matrix-multiplication-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/d5fee5b55a64e47f1b5724ec39adf171/03-matrix-multiplication.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">03-matrix-multiplication.py</span></code></a></p>

2
v1.1.2/getting-started/tutorials/04-low-memory-dropout.html
View File

@@ -371,7 +371,7 @@ to explore the <cite>triton/language/random</cite> folder!</p>
<dd><p>Nitish Srivastava and Geoffrey Hinton and Alex Krizhevsky and Ilya Sutskever and Ruslan Salakhutdinov, “Dropout: A Simple Way to Prevent Neural Networks from Overfitting”, JMLR 2014</p>
</dd>
</dl>
<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 0 minutes 0.011 seconds)</p>
<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 0 minutes 0.110 seconds)</p>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-getting-started-tutorials-04-low-memory-dropout-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/c9aed78977a4c05741d675a38dde3d7d/04-low-memory-dropout.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">04-low-memory-dropout.py</span></code></a></p>

62
v1.1.2/getting-started/tutorials/05-layer-norm.html
View File

@@ -194,36 +194,36 @@ to download the full example code</p>
<p class="sphx-glr-script-out">Out:</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>layer-norm-backward:
N Triton Torch Apex
0 1024.0 303.407414 98.698793 311.088617
1 1536.0 347.773587 133.083026 341.333333
2 2048.0 416.542360 157.538467 332.108094
3 2560.0 451.764698 181.238943 328.556154
4 3072.0 508.468972 190.511624 320.556515
5 3584.0 540.981122 206.769233 308.301075
6 4096.0 558.545450 219.919464 298.796351
7 4608.0 489.345125 231.364016 286.507772
8 5120.0 520.677950 242.366855 285.767451
9 5632.0 534.260858 243.545956 291.310338
10 6144.0 544.118087 249.925419 286.879370
11 6656.0 532.479975 254.775119 285.767438
12 7168.0 515.065851 252.988236 277.024148
13 7680.0 488.912481 265.590783 283.569230
14 8192.0 464.794337 257.677592 277.303250
15 8704.0 408.798442 266.448988 284.212242
16 9216.0 422.106891 271.391419 289.129410
17 9728.0 430.760152 279.272720 288.237038
18 10240.0 438.074849 286.433562 289.129408
19 10752.0 426.525614 245.760009 289.291486
20 11264.0 427.071098 244.426754 285.465683
21 11776.0 418.082825 248.569911 288.097854
22 12288.0 416.542386 253.578674 293.737063
23 12800.0 412.348979 253.047766 288.993430
24 13312.0 410.125805 251.367424 288.607034
25 13824.0 403.130022 256.197690 291.031592
26 14336.0 395.021816 255.051144 288.402346
27 14848.0 384.829370 256.737757 288.310684
28 15360.0 376.547496 257.430175 287.550706
29 15872.0 369.832994 260.731015 289.899545
0 1024.0 307.200008 99.497980 315.076934
1 1536.0 347.773587 134.050910 344.523365
2 2048.0 423.724127 159.067963 323.368435
3 2560.0 461.954908 182.314537 325.079368
4 3072.0 515.580429 191.501303 319.168834
5 3584.0 551.384634 207.768111 309.410081
6 4096.0 564.965515 220.907859 301.546004
7 4608.0 498.162157 232.825259 287.251954
8 5120.0 529.655159 243.809526 286.433562
9 5632.0 540.671974 244.426754 291.939522
10 6144.0 550.208948 251.202731 287.438593
11 6656.0 534.260858 255.590406 286.793541
12 7168.0 515.065851 253.734520 277.470965
13 7680.0 490.212752 266.743841 284.884090
14 8192.0 464.794337 258.354805 278.087683
15 8704.0 416.127506 267.815384 285.767450
16 9216.0 430.319054 272.059034 289.887291
17 9728.0 438.033784 279.942444 288.950501
18 10240.0 446.836366 287.438599 290.496460
19 10752.0 430.079980 246.699797 289.941565
20 11264.0 430.471331 245.313973 286.069848
21 11776.0 421.198220 249.447482 288.686414
22 12288.0 418.314886 254.673582 294.617366
23 12800.0 414.016170 254.094291 289.538159
24 13312.0 412.242569 252.559690 289.129403
25 13824.0 405.098897 256.991469 291.799461
26 14336.0 396.387109 256.000002 289.129416
27 14848.0 386.918555 257.665934 289.012175
28 15360.0 376.932517 258.332158 286.656296
29 15872.0 369.832994 261.986243 290.784741
</pre></div>
</div>
<div class="line-block">
@@ -477,7 +477,7 @@ to download the full example code</p>
<span class="n">bench_layer_norm</span><span class="o">.</span><span class="n">run</span><span class="p">(</span><span class="n">save_path</span><span class="o">=</span><span class="s1">&#39;.&#39;</span><span class="p">,</span> <span class="n">print_data</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
</pre></div>
</div>
<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 13.777 seconds)</p>
<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 12.361 seconds)</p>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-getting-started-tutorials-05-layer-norm-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/935c0dd0fbeb4b2e69588471cbb2d4b2/05-layer-norm.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">05-layer-norm.py</span></code></a></p>

12
v1.1.2/getting-started/tutorials/sg_execution_times.html
View File

@@ -174,7 +174,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-getting-started-tutorials-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline"></a></h1>
<p><strong>12:46.783</strong> total execution time for <strong>getting-started_tutorials</strong> files:</p>
<p><strong>12:56.474</strong> total execution time for <strong>getting-started_tutorials</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 85%" />
@@ -183,23 +183,23 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="03-matrix-multiplication.html#sphx-glr-getting-started-tutorials-03-matrix-multiplication-py"><span class="std std-ref">Matrix Multiplication</span></a> (<code class="docutils literal notranslate"><span class="pre">03-matrix-multiplication.py</span></code>)</p></td>
<td><p>05:23.021</p></td>
<td><p>05:34.991</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="02-fused-softmax.html#sphx-glr-getting-started-tutorials-02-fused-softmax-py"><span class="std std-ref">Fused Softmax</span></a> (<code class="docutils literal notranslate"><span class="pre">02-fused-softmax.py</span></code>)</p></td>
<td><p>03:25.535</p></td>
<td><p>03:24.321</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="05-layer-norm.html#sphx-glr-getting-started-tutorials-05-layer-norm-py"><span class="std std-ref">Layer Normalization</span></a> (<code class="docutils literal notranslate"><span class="pre">05-layer-norm.py</span></code>)</p></td>
<td><p>02:13.777</p></td>
<td><p>02:12.361</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="01-vector-add.html#sphx-glr-getting-started-tutorials-01-vector-add-py"><span class="std std-ref">Vector Addition</span></a> (<code class="docutils literal notranslate"><span class="pre">01-vector-add.py</span></code>)</p></td>
<td><p>01:44.439</p></td>
<td><p>01:44.691</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="04-low-memory-dropout.html#sphx-glr-getting-started-tutorials-04-low-memory-dropout-py"><span class="std std-ref">Low-Memory Dropout</span></a> (<code class="docutils literal notranslate"><span class="pre">04-low-memory-dropout.py</span></code>)</p></td>
<td><p>00:00.011</p></td>
<td><p>00:00.110</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>