Merge branch 'master' into resolve_inputs_fix

Author: narendasan

docs/_cpp_api/function_torch__tensorrt_8h_1a710df824a7718b440e4bc17bf9693cef.html

@@ -635,6 +635,21 @@ <h2 id="function-documentation">
            <p>
             Takes a pre-built serialized TensorRT engine and embeds it in a TorchScript module. Registers execution of the engine as the forward method of the module Forward is defined as: forward(Tensor[]) -&gt; Tensor[]
            </p>
+           <p>
+            TensorRT bindings must have names with the following format:
+           </p>
+           <ul class="simple">
+            <li>
+             <p>
+              [symbol].[index in input / output array] ex.
+             </p>
+            </li>
+            <li>
+             <p>
+              [x.0, x.1, x.2] -&gt; [y.0]
+             </p>
+            </li>
+           </ul>
            <dl class="field-list simple">
             <dt class="field-odd">
              Parameters
@@ -670,7 +685,7 @@ <h2 id="function-documentation">
             </dt>
             <dd class="field-even">
              <p>
-              : A new module trageting a TensorRT engine
+              : A new module targeting a TensorRT engine
              </p>
             </dd>
            </dl>

docs/_notebooks/CitriNet-example.html

@@ -857,7 +857,7 @@
         </div>
        </div>
        <p>
-        <img alt="20d57019de6243a68e76654f4ea8f8ae" src="http://developer.download.nvidia.com/compute/machine-learning/frameworks/nvidia_logo.png"/>
+        <img alt="c61fa285604242db814bfb71fcb3f44a" src="http://developer.download.nvidia.com/compute/machine-learning/frameworks/nvidia_logo.png"/>
        </p>
        <section id="Torch-TensorRT-Getting-Started---CitriNet">
         <h1 id="notebooks-citrinet-example--page-root">

docs/_notebooks/EfficientNet-example.html

@@ -857,7 +857,7 @@
         </div>
        </div>
        <p>
-        <img alt="8ef80a67a46b4935a6252c9d572a57b8" src="http://developer.download.nvidia.com/compute/machine-learning/frameworks/nvidia_logo.png"/>
+        <img alt="1b913ec657a24201a6c8f2c361875572" src="http://developer.download.nvidia.com/compute/machine-learning/frameworks/nvidia_logo.png"/>
        </p>
        <section id="Torch-TensorRT-Getting-Started---EfficientNet-B0">
         <h1 id="notebooks-efficientnet-example--page-root">

docs/_notebooks/Hugging-Face-BERT.html

@@ -809,7 +809,7 @@
         </div>
        </div>
        <p>
-        <img alt="2c52902157e54c8e8068349075e9eeff" src="https://developer.download.nvidia.com/tesla/notebook_assets/nv_logo_torch_trt_resnet_notebook.png"/>
+        <img alt="9789ac3c1cdf4cef840d6171a4052c75" src="https://developer.download.nvidia.com/tesla/notebook_assets/nv_logo_torch_trt_resnet_notebook.png"/>
        </p>
        <section id="Masked-Language-Modeling-(MLM)-with-Hugging-Face-BERT-Transformer">
         <h1 id="notebooks-hugging-face-bert--page-root">

docs/_notebooks/Resnet50-example.html

@@ -857,7 +857,7 @@
         </div>
        </div>
        <p>
-        <img alt="9cbd1c2fc7c541b5b2afa929886f578b" src="https://developer.download.nvidia.com/tesla/notebook_assets/nv_logo_torch_trt_resnet_notebook.png"/>
+        <img alt="a1c22a88cb274311833320666e204a51" src="https://developer.download.nvidia.com/tesla/notebook_assets/nv_logo_torch_trt_resnet_notebook.png"/>
        </p>
        <section id="Torch-TensorRT-Getting-Started---ResNet-50">
         <h1 id="notebooks-resnet50-example--page-root">

docs/_notebooks/lenet-getting-started.html

@@ -895,7 +895,7 @@
         </div>
        </div>
        <p>
-        <img alt="ff2c118facde47ba9f8b3b3129d8fb55" src="http://developer.download.nvidia.com/compute/machine-learning/frameworks/nvidia_logo.png"/>
+        <img alt="3f6a6bf3756a4dc996a1bfeea83e15e1" src="http://developer.download.nvidia.com/compute/machine-learning/frameworks/nvidia_logo.png"/>
        </p>
        <section id="Torch-TensorRT-Getting-Started---LeNet">
         <h1 id="notebooks-lenet-getting-started--page-root">

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

@@ -915,7 +915,7 @@
         </div>
        </div>
        <p>
-        <img alt="76b4640a03564b6c8e5f8ec237a17f05" src="https://developer.download.nvidia.com/tesla/notebook_assets/nv_logo_torch_trt_ssd_notebook.png"/>
+        <img alt="aac56f28a5e54218ae3d219392f48c65" src="https://developer.download.nvidia.com/tesla/notebook_assets/nv_logo_torch_trt_ssd_notebook.png"/>
        </p>
        <section id="Object-Detection-with-Torch-TensorRT-(SSD)">
         <h1 id="notebooks-ssd-object-detection-demo--page-root">

docs/_notebooks/vgg-qat.html

@@ -922,7 +922,7 @@ <h2 id="Overview">
          </div>
          <p>
           ## 2. VGG16 Overview ### Very Deep Convolutional Networks for Large-Scale Image Recognition VGG is one of the earliest family of image classification networks that first used small (3x3) convolution filters and achieved significant improvements on ImageNet recognition challenge. The network architecture looks as follows
-          <img alt="c96789a7ffcd49a291298042bad651d4" src="https://neurohive.io/wp-content/uploads/2018/11/vgg16-1-e1542731207177.png"/>
+          <img alt="65356fb264d64a9a9295dea65f568939" src="https://neurohive.io/wp-content/uploads/2018/11/vgg16-1-e1542731207177.png"/>
          </p>
          <p>
           ## 3. Training a baseline VGG16 model We train VGG16 on CIFAR10 dataset. Define training and testing datasets and dataloaders. This will download the CIFAR 10 data in your

docs/py_api/ts.html

@@ -1851,6 +1851,24 @@ <h2 id="functions">
              </p>
             </div>
            </blockquote>
+           <dl>
+            <dt>
+             TensorRT bindings must have names with the following format:
+            </dt>
+            <dd>
+             <ul class="simple">
+              <li>
+               <p>
+                [symbol].[index in input / output array]
+               </p>
+              </li>
+             </ul>
+             <p>
+              ex.
+- [x.0, x.1, x.2] -&gt; [y.0]
+             </p>
+            </dd>
+           </dl>
            <p>
             Module can be save with engine embedded with torch.jit.save and moved / loaded according to torch_tensorrt portability rules
            </p>
@@ -2066,7 +2084,7 @@ <h2 id="functions">
               at
              </span>
              <span class="pre">
-              0x7f97985ef6f0&gt;
+              0x7fdeab7877b0&gt;
              </span>
             </span>
            </span>