Skip to content

Commit 5ea5b24

Browse files
author
TRTorch Github Bot
committed
docs: [Automated] Regenerating documenation from
Signed-off-by: TRTorch Github Bot <[email protected]>
1 parent 19d301b commit 5ea5b24

8 files changed

+54
-106
lines changed

‎docs/_cpp_api/program_listing_file_cpp_api_include_trtorch_trtorch.h.html

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -410,7 +410,7 @@ <h1 id="cpp-api-program-listing-file-cpp-api-include-trtorch-trtorch-h--page-roo
410410
<span class="p">}</span> <span class="c1">// namespace torch</span>
411411

412412
<span class="k">namespace</span> <span class="n">c10</span> <span class="p">{</span>
413-
<span class="k">enum</span> <span class="k">class</span> <span class="nc">DeviceType</span> <span class="o">:</span> <span class="kt">int16_t</span><span class="p">;</span>
413+
<span class="k">enum</span> <span class="k">class</span> <span class="nc">DeviceType</span> <span class="o">:</span> <span class="kt">int8_t</span><span class="p">;</span>
414414
<span class="k">enum</span> <span class="k">class</span> <span class="nc">ScalarType</span> <span class="o">:</span> <span class="kt">int8_t</span><span class="p">;</span>
415415
<span class="k">template</span> <span class="o">&lt;</span><span class="k">class</span><span class="o">&gt;</span>
416416
<span class="k">class</span> <span class="nc">ArrayRef</span><span class="p">;</span>

‎docs/_notebooks/Resnet50-example.html

Lines changed: 20 additions & 40 deletions
Original file line numberDiff line numberDiff line change
@@ -405,7 +405,7 @@
405405
</a>
406406
</li>
407407
<li class="md-nav__item">
408-
<a class="md-nav__link" href="#What's-next">
408+
<a class="md-nav__link" href="#What’s-next">
409409
What’s next
410410
</a>
411411
</li>
@@ -690,7 +690,7 @@
690690
</div>
691691
</div>
692692
<p>
693-
<img alt="0904ed52551d45acab1d028795d47df1" src="http://developer.download.nvidia.com/compute/machine-learning/frameworks/nvidia_logo.png"/>
693+
<img alt="99f904f675e04b9aa332ef722639d55e" src="http://developer.download.nvidia.com/compute/machine-learning/frameworks/nvidia_logo.png"/>
694694
</p>
695695
<h1 id="notebooks-resnet50-example--page-root">
696696
TRTorch Getting Started - ResNet 50
@@ -840,13 +840,9 @@ <h2 id="Content">
840840
</div>
841841
</div>
842842
</div>
843-
<blockquote>
844-
<div>
845-
<p>
846-
## 1. Requirements
847-
</p>
848-
</div>
849-
</blockquote>
843+
<p>
844+
## 1. Requirements
845+
</p>
850846
<p>
851847
Follow the steps in
852848
<code class="docutils literal notranslate">
@@ -856,13 +852,9 @@ <h2 id="Content">
856852
</code>
857853
to prepare a Docker container, within which you can run this notebook.
858854
</p>
859-
<blockquote>
860-
<div>
861-
<p>
862-
## 2. ResNet-50 Overview
863-
</p>
864-
</div>
865-
</blockquote>
855+
<p>
856+
## 2. ResNet-50 Overview
857+
</p>
866858
<p>
867859
PyTorch has a model repository called the PyTorch Hub, which is a source for high quality implementations of common models. We can get our ResNet-50 model from there pretrained on ImageNet.
868860
</p>
@@ -1101,7 +1093,7 @@ <h3 id="Model-Description">
11011093
</div>
11021094
</div>
11031095
<p>
1104-
All pre-trained models expect input images normalized in the same way, i.e. mini-batches of 3-channel RGB images of shape
1096+
All pre-trained models expect input images normalized in the same way, i.e. mini-batches of 3-channel RGB images of shape
11051097
<code class="docutils literal notranslate">
11061098
<span class="pre">
11071099
(3
@@ -1569,13 +1561,9 @@ <h3 id="Benchmark-utility">
15691561
</div>
15701562
</div>
15711563
</div>
1572-
<blockquote>
1573-
<div>
1574-
<p>
1575-
## 3. Creating TorchScript modules
1576-
</p>
1577-
</div>
1578-
</blockquote>
1564+
<p>
1565+
## 3. Creating TorchScript modules
1566+
</p>
15791567
<p>
15801568
To compile with TRTorch, the model must first be in
15811569
<strong>
@@ -1694,13 +1682,9 @@ <h2 id="Tracing">
16941682
</div>
16951683
</div>
16961684
</div>
1697-
<blockquote>
1698-
<div>
1699-
<p>
1700-
## 4. Compiling with TRTorch
1701-
</p>
1702-
</div>
1703-
</blockquote>
1685+
<p>
1686+
## 4. Compiling with TRTorch
1687+
</p>
17041688
<p>
17051689
TorchScript modules behave just like normal PyTorch modules and are intercompatible. From TorchScript we can now compile a TensorRT based module. This module will still be implemented in TorchScript but all the computation will be done in TensorRT.
17061690
</p>
@@ -1853,13 +1837,9 @@ <h3 id="FP16-(half-precision)">
18531837
</div>
18541838
</div>
18551839
</div>
1856-
<blockquote>
1857-
<div>
1858-
<p>
1859-
## 5. Conclusion
1860-
</p>
1861-
</div>
1862-
</blockquote>
1840+
<p>
1841+
## 5. Conclusion
1842+
</p>
18631843
<p>
18641844
In this notebook, we have walked through the complete process of compiling TorchScript models with TRTorch for ResNet-50 model and test the performance impact of the optimization. With TRTorch, we observe a speedup of
18651845
<strong>
@@ -1871,9 +1851,9 @@ <h3 id="FP16-(half-precision)">
18711851
</strong>
18721852
with FP16.
18731853
</p>
1874-
<h3 id="What's-next">
1854+
<h3 id="What’s-next">
18751855
What’s next
1876-
<a class="headerlink" href="#What's-next" title="Permalink to this headline">
1856+
<a class="headerlink" href="#What’s-next" title="Permalink to this headline">
18771857
18781858
</a>
18791859
</h3>

‎docs/_notebooks/lenet-getting-started.html

Lines changed: 14 additions & 26 deletions
Original file line numberDiff line numberDiff line change
@@ -339,7 +339,7 @@
339339
<nav class="md-nav">
340340
<ul class="md-nav__list">
341341
<li class="md-nav__item">
342-
<a class="md-nav__link" href="#What's-next">
342+
<a class="md-nav__link" href="#What’s-next">
343343
What’s next
344344
</a>
345345
</li>
@@ -499,7 +499,7 @@
499499
<nav class="md-nav">
500500
<ul class="md-nav__list">
501501
<li class="md-nav__item">
502-
<a class="md-nav__link" href="#What's-next">
502+
<a class="md-nav__link" href="#What’s-next">
503503
What’s next
504504
</a>
505505
</li>
@@ -784,7 +784,7 @@
784784
</div>
785785
</div>
786786
<p>
787-
<img alt="eb83af0673044a89b9357bb998acd927" src="http://developer.download.nvidia.com/compute/machine-learning/frameworks/nvidia_logo.png"/>
787+
<img alt="e099c76826334bc58700d7d079b4de7c" src="http://developer.download.nvidia.com/compute/machine-learning/frameworks/nvidia_logo.png"/>
788788
</p>
789789
<h1 id="notebooks-lenet-getting-started--page-root">
790790
TRTorch Getting Started - LeNet
@@ -848,13 +848,9 @@ <h2 id="Content">
848848
</p>
849849
</li>
850850
</ol>
851-
<blockquote>
852-
<div>
853-
<p>
854-
## 1. Requirements
855-
</p>
856-
</div>
857-
</blockquote>
851+
<p>
852+
## 1. Requirements
853+
</p>
858854
<p>
859855
Follow the steps in
860856
<code class="docutils literal notranslate">
@@ -864,13 +860,9 @@ <h2 id="Content">
864860
</code>
865861
to prepare a Docker container, within which you can run this notebook.
866862
</p>
867-
<blockquote>
868-
<div>
869-
<p>
870-
## 2. Creating TorchScript modules
871-
</p>
872-
</div>
873-
</blockquote>
863+
<p>
864+
## 2. Creating TorchScript modules
865+
</p>
874866
<p>
875867
Here we create two submodules for a feature extractor and a classifier and stitch them together in a single LeNet module. In this case this is overkill but modules give us granular control over our program including where we decide to optimize and where we don’t. It is also the unit that the TorchScript compiler operates on. So you can decide to only convert/optimize the feature extractor and leave the classifier in standard PyTorch or you can convert the whole thing. When compiling your module
876868
to TorchScript, there are two paths: Tracing and Scripting.
@@ -1272,13 +1264,9 @@ <h3 id="Scripting">
12721264
</div>
12731265
</div>
12741266
</div>
1275-
<blockquote>
1276-
<div>
1277-
<p>
1278-
## 3. Compiling with TRTorch
1279-
</p>
1280-
</div>
1281-
</blockquote>
1267+
<p>
1268+
## 3. Compiling with TRTorch
1269+
</p>
12821270
<h3 id="TorchScript-traced-model">
12831271
TorchScript traced model
12841272
<a class="headerlink" href="#TorchScript-traced-model" title="Permalink to this headline">
@@ -1458,9 +1446,9 @@ <h2 id="Conclusion">
14581446
<p>
14591447
In this notebook, we have walked through the complete process of compiling TorchScript models with TRTorch and test the performance impact of the optimization.
14601448
</p>
1461-
<h3 id="What's-next">
1449+
<h3 id="What’s-next">
14621450
What’s next
1463-
<a class="headerlink" href="#What's-next" title="Permalink to this headline">
1451+
<a class="headerlink" href="#What’s-next" title="Permalink to this headline">
14641452
14651453
</a>
14661454
</h3>

‎docs/_notebooks/ssd-object-detection-demo.html

Lines changed: 16 additions & 36 deletions
Original file line numberDiff line numberDiff line change
@@ -804,7 +804,7 @@
804804
</div>
805805
</div>
806806
<p>
807-
<img alt="092b8936ef17484f92965a48c4cdea98" src="http://developer.download.nvidia.com/compute/machine-learning/frameworks/nvidia_logo.png"/>
807+
<img alt="7688c88a8c154b328dac21c581f2777a" src="http://developer.download.nvidia.com/compute/machine-learning/frameworks/nvidia_logo.png"/>
808808
</p>
809809
<h1 id="notebooks-ssd-object-detection-demo--page-root">
810810
Object Detection with TRTorch (SSD)
@@ -902,13 +902,9 @@ <h2 id="Contents">
902902
</li>
903903
</ol>
904904
<hr class="docutils"/>
905-
<blockquote>
906-
<div>
907-
<p>
908-
## 1. Requirements
909-
</p>
910-
</div>
911-
</blockquote>
905+
<p>
906+
## 1. Requirements
907+
</p>
912908
<p>
913909
Follow the steps in
914910
<code class="docutils literal notranslate">
@@ -940,13 +936,9 @@ <h2 id="Contents">
940936
</div>
941937
</div>
942938
<hr class="docutils"/>
943-
<blockquote>
944-
<div>
945-
<p>
946-
## 2. SSD
947-
</p>
948-
</div>
949-
</blockquote>
939+
<p>
940+
## 2. SSD
941+
</p>
950942
<h3 id="Single-Shot-MultiBox-Detector-model-for-object-detection">
951943
Single Shot MultiBox Detector model for object detection
952944
<a class="headerlink" href="#Single-Shot-MultiBox-Detector-model-for-object-detection" title="Permalink to this headline">
@@ -1386,13 +1378,9 @@ <h3 id="Benchmark-utility">
13861378
</div>
13871379
</div>
13881380
<hr class="docutils"/>
1389-
<blockquote>
1390-
<div>
1391-
<p>
1392-
## 3. Creating TorchScript modules
1393-
</p>
1394-
</div>
1395-
</blockquote>
1381+
<p>
1382+
## 3. Creating TorchScript modules
1383+
</p>
13961384
<p>
13971385
To compile with TRTorch, the model must first be in
13981386
<strong>
@@ -1490,13 +1478,9 @@ <h3 id="Benchmark-utility">
14901478
</div>
14911479
</div>
14921480
<hr class="docutils"/>
1493-
<blockquote>
1494-
<div>
1495-
<p>
1496-
## 4. Compiling with TRTorch TorchScript modules behave just like normal PyTorch modules and are intercompatible. From TorchScript we can now compile a TensorRT based module. This module will still be implemented in TorchScript but all the computation will be done in TensorRT.
1497-
</p>
1498-
</div>
1499-
</blockquote>
1481+
<p>
1482+
## 4. Compiling with TRTorch TorchScript modules behave just like normal PyTorch modules and are intercompatible. From TorchScript we can now compile a TensorRT based module. This module will still be implemented in TorchScript but all the computation will be done in TensorRT.
1483+
</p>
15001484
<div class="nbinput nblast docutils container">
15011485
<div class="prompt highlight-none notranslate">
15021486
<div class="highlight">
@@ -1521,13 +1505,9 @@ <h3 id="Benchmark-utility">
15211505
</div>
15221506
</div>
15231507
<hr class="docutils"/>
1524-
<blockquote>
1525-
<div>
1526-
<p>
1527-
## 5. Running Inference
1528-
</p>
1529-
</div>
1530-
</blockquote>
1508+
<p>
1509+
## 5. Running Inference
1510+
</p>
15311511
<p>
15321512
Next, we run object detection
15331513
</p>

‎docs/_sources/_cpp_api/program_listing_file_cpp_api_include_trtorch_trtorch.h.rst.txt

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -35,7 +35,7 @@ Program Listing for File trtorch.h
3535
} // namespace torch
3636
3737
namespace c10 {
38-
enum class DeviceType : int16_t;
38+
enum class DeviceType : int8_t;
3939
enum class ScalarType : int8_t;
4040
template <class>
4141
class ArrayRef;

‎docs/objects.inv

9 Bytes
Binary file not shown.

‎docs/py_api/trtorch.html

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1011,7 +1011,7 @@ <h2 id="functions">
10111011
<span class="sig-paren">
10121012
)
10131013
</span>
1014-
→ &lt;torch._C.ScriptClass object at 0x7fb23f032618&gt;
1014+
→ &lt;torch._C.ScriptClass object at 0x7f3c597348f0&gt;
10151015
<a class="headerlink" href="#trtorch.TensorRTCompileSpec" title="Permalink to this definition">
10161016
10171017
</a>

‎docs/searchindex.js

Lines changed: 1 addition & 1 deletion
Some generated files are not rendered by default. Learn more about customizing how changed files appear on GitHub.

0 commit comments

Comments
 (0)