docs: Document the tuple lowering pass

Signed-off-by: Naren Dasan <[email protected]>
Signed-off-by: Naren Dasan <[email protected]>

Author: narendasan

docs/_sources/contributors/lowering.rst.txt

@@ -84,6 +84,19 @@ all attributes accesses are replaced with explicit inputs of the graph
 graph are the trainable parameters used in this method. The remaining inputs
 are the true inputs to the function.
 
+Lower Tuples
+***************************************
+
+    `torch/csrc/jit/passes/lower_tuples.h <https://github.com/pytorch/pytorch/blob/master/torch/csrc/jit/passes/lower_tuples.h>`_
+
+* ``LowerSimpleTuples``:
+
+Removes tuples where TupleConstruct and TupleUnpack are matched but leaves tuples in place across if statements, loops, and as inputs/outputs
+
+* ``LowerAllTuples``:
+
+Removes _all_ tuples and raises an error if some cannot be removed, this is used by ONNX to ensure there are not tuples before conversion, but will not work on graphs whose inputs contain tuples.
+
 Remove Dropout
 ***************************************
 

docs/contributors/lowering.html

@@ -522,6 +522,11 @@
                   Lower Graph
                  </a>
                 </li>
+                <li class="md-nav__item">
+                 <a class="md-nav__link" href="#lower-tuples">
+                  Lower Tuples
+                 </a>
+                </li>
                 <li class="md-nav__item">
                  <a class="md-nav__link" href="#remove-dropout">
                   Remove Dropout
@@ -764,6 +769,51 @@ <h3 id="lower-graph">
 graph are the trainable parameters used in this method. The remaining inputs
 are the true inputs to the function.
        </p>
+       <h3 id="lower-tuples">
+        Lower Tuples
+        <a class="headerlink" href="#lower-tuples" title="Permalink to this headline">
+         ¶
+        </a>
+       </h3>
+       <blockquote>
+        <div>
+         <p>
+          <a class="reference external" href="https://github.com/pytorch/pytorch/blob/master/torch/csrc/jit/passes/lower_tuples.h">
+           torch/csrc/jit/passes/lower_tuples.h
+          </a>
+         </p>
+        </div>
+       </blockquote>
+       <ul class="simple">
+        <li>
+         <p>
+          <code class="docutils literal notranslate">
+           <span class="pre">
+            LowerSimpleTuples
+           </span>
+          </code>
+          :
+         </p>
+        </li>
+       </ul>
+       <p>
+        Removes tuples where TupleConstruct and TupleUnpack are matched but leaves tuples in place across if statements, loops, and as inputs/outputs
+       </p>
+       <ul class="simple">
+        <li>
+         <p>
+          <code class="docutils literal notranslate">
+           <span class="pre">
+            LowerAllTuples
+           </span>
+          </code>
+          :
+         </p>
+        </li>
+       </ul>
+       <p>
+        Removes _all_ tuples and raises an error if some cannot be removed, this is used by ONNX to ensure there are not tuples before conversion, but will not work on graphs whose inputs contain tuples.
+       </p>
        <h3 id="remove-dropout">
         Remove Dropout
         <a class="headerlink" href="#remove-dropout" title="Permalink to this headline">

docs/searchindex.js

@@ -1 +1 @@
-<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9"><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/class_view_hierarchy.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/classtrtorch_1_1ExtraInfo_1_1DataType.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/classtrtorch_1_1ExtraInfo_1_1DeviceType.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/classtrtorch_1_1ptq_1_1Int8CacheCalibrator.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/classtrtorch_1_1ptq_1_1Int8Calibrator.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/define_macros_8h_1a18d295a837ac71add5578860b55e5502.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/define_macros_8h_1a20c1fbeb21757871c52299dc52351b5f.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/define_macros_8h_1a25ee153c325dfc7466a33cbd5c1ff055.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/define_macros_8h_1a48d6029a45583a06848891cb0e86f7ba.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/define_macros_8h_1a71b02dddfabe869498ad5a88e11c440f.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/define_macros_8h_1a9d31d0569348d109b1b069b972dd143e.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/define_macros_8h_1abe87b341f562fd1cf40b7672e4d759da.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/define_macros_8h_1ae1c56ab8a40af292a9a4964651524d84.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/dir_cpp.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/dir_cpp_api.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/dir_cpp_api_include.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/dir_cpp_api_include_trtorch.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/enum_logging_8h_1a5f612ff2f783ff4fbe89d168f0d817d4.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/file_cpp_api_include_trtorch_logging.h.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/file_cpp_api_include_trtorch_macros.h.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/file_cpp_api_include_trtorch_ptq.h.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/file_cpp_api_include_trtorch_trtorch.h.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/file_view_hierarchy.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/function_logging_8h_1a118d65b179defff7fff279eb9cd126cb.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/function_logging_8h_1a396a688110397538f8b3fb7dfdaf38bb.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/function_logging_8h_1a9b420280bfacc016d7e36a5704021949.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/function_logging_8h_1aa533955a2b908db9e5df5acdfa24715f.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/function_logging_8h_1abc57d473f3af292551dee8b9c78373ad.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/function_logging_8h_1adf5435f0dbb09c0d931a1b851847236b.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/function_logging_8h_1aef44b69c62af7cf2edc8875a9506641a.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/function_trtorch_8h_1a2cf17d43ba9117b3b4d652744b4f0447.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/function_trtorch_8h_1a4422781719d7befedb364cacd91c6247.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/function_trtorch_8h_1a536bba54b70e44554099d23fa3d7e804.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/function_trtorch_8h_1a5f33b142bc2f3f2aaf462270b3ad7e31.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/function_trtorch_8h_1a726f6e7091b6b7be45b5a4275b2ffb10.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/function_trtorch_8h_1ab01696cfe08b6a5293c55935a9713c25.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/function_trtorch_8h_1ae38897d1ca4438227c970029d0f76fb5.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/namespace_trtorch.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/namespace_trtorch__logging.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/namespace_trtorch__ptq.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/program_listing_file_cpp_api_include_trtorch_logging.h.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/program_listing_file_cpp_api_include_trtorch_macros.h.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/program_listing_file_cpp_api_include_trtorch_ptq.h.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/program_listing_file_cpp_api_include_trtorch_trtorch.h.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/structtrtorch_1_1ExtraInfo.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/structtrtorch_1_1ExtraInfo_1_1InputRange.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/trtorch_cpp.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/unabridged_api.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/unabridged_orphan.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/index.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/tutorials/installation.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/genindex.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/py-modindex.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/search.html</loc></url></urlset>
+<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9"><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/class_view_hierarchy.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/classtrtorch_1_1ExtraInfo_1_1DataType.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/classtrtorch_1_1ExtraInfo_1_1DeviceType.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/classtrtorch_1_1ptq_1_1Int8CacheCalibrator.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/classtrtorch_1_1ptq_1_1Int8Calibrator.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/define_macros_8h_1a18d295a837ac71add5578860b55e5502.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/define_macros_8h_1a20c1fbeb21757871c52299dc52351b5f.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/define_macros_8h_1a25ee153c325dfc7466a33cbd5c1ff055.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/define_macros_8h_1a48d6029a45583a06848891cb0e86f7ba.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/define_macros_8h_1a71b02dddfabe869498ad5a88e11c440f.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/define_macros_8h_1a9d31d0569348d109b1b069b972dd143e.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/define_macros_8h_1abe87b341f562fd1cf40b7672e4d759da.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/define_macros_8h_1ae1c56ab8a40af292a9a4964651524d84.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/dir_cpp.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/dir_cpp_api.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/dir_cpp_api_include.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/dir_cpp_api_include_trtorch.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/enum_logging_8h_1a5f612ff2f783ff4fbe89d168f0d817d4.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/file_cpp_api_include_trtorch_logging.h.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/file_cpp_api_include_trtorch_macros.h.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/file_cpp_api_include_trtorch_ptq.h.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/file_cpp_api_include_trtorch_trtorch.h.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/file_view_hierarchy.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/function_logging_8h_1a118d65b179defff7fff279eb9cd126cb.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/function_logging_8h_1a396a688110397538f8b3fb7dfdaf38bb.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/function_logging_8h_1a9b420280bfacc016d7e36a5704021949.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/function_logging_8h_1aa533955a2b908db9e5df5acdfa24715f.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/function_logging_8h_1abc57d473f3af292551dee8b9c78373ad.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/function_logging_8h_1adf5435f0dbb09c0d931a1b851847236b.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/function_logging_8h_1aef44b69c62af7cf2edc8875a9506641a.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/function_trtorch_8h_1a2cf17d43ba9117b3b4d652744b4f0447.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/function_trtorch_8h_1a4422781719d7befedb364cacd91c6247.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/function_trtorch_8h_1a536bba54b70e44554099d23fa3d7e804.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/function_trtorch_8h_1a5f33b142bc2f3f2aaf462270b3ad7e31.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/function_trtorch_8h_1a726f6e7091b6b7be45b5a4275b2ffb10.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/function_trtorch_8h_1ab01696cfe08b6a5293c55935a9713c25.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/function_trtorch_8h_1ae38897d1ca4438227c970029d0f76fb5.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/namespace_trtorch.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/namespace_trtorch__logging.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/namespace_trtorch__ptq.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/program_listing_file_cpp_api_include_trtorch_logging.h.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/program_listing_file_cpp_api_include_trtorch_macros.h.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/program_listing_file_cpp_api_include_trtorch_ptq.h.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/program_listing_file_cpp_api_include_trtorch_trtorch.h.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/structtrtorch_1_1ExtraInfo.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/structtrtorch_1_1ExtraInfo_1_1InputRange.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/trtorch_cpp.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/unabridged_api.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/_cpp_api/unabridged_orphan.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/contributors/lowering.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/contributors/phases.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/contributors/system_overview.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/contributors/useful_links.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/index.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/py_api/logging.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/py_api/trtorch.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/tutorials/getting_started.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/tutorials/installation.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/genindex.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/py-modindex.html</loc></url><url><loc>https://nvidia.github.io/TRTorch/search.html</loc></url></urlset>

docs/sitemap.xml

@@ -937,10 +937,10 @@ <h2 id="working-with-torchscript-in-c">
 <span class="cp">#include</span> <span class="cpf">&lt;memory&gt;</span><span class="cp"></span>
 
 <span class="kt">int</span> <span class="nf">main</span><span class="p">(</span><span class="kt">int</span> <span class="n">argc</span><span class="p">,</span> <span class="k">const</span> <span class="kt">char</span><span class="o">*</span> <span class="n">argv</span><span class="p">[])</span> <span class="p">{</span>
-    <span class="n">torch</span><span class="o">::</span><span class="n">jit</span><span class="o">::</span><span class="n">Module</span> <span class="n">module</span><span class="p">;</span>
+    <span class="n">torch</span><span class="o">::</span><span class="n">jit</span><span class="o">::</span><span class="n">Module</span> <span class="k">module</span><span class="p">;</span>
     <span class="k">try</span> <span class="p">{</span>
         <span class="c1">// Deserialize the ScriptModule from a file using torch::jit::load().</span>
-        <span class="n">module</span> <span class="o">=</span> <span class="n">torch</span><span class="o">::</span><span class="n">jit</span><span class="o">::</span><span class="n">load</span><span class="p">(</span><span class="s">"&lt;PATH TO SAVED TS MOD&gt;"</span><span class="p">);</span>
+        <span class="k">module</span> <span class="o">=</span> <span class="n">torch</span><span class="o">::</span><span class="n">jit</span><span class="o">::</span><span class="n">load</span><span class="p">(</span><span class="s">"&lt;PATH TO SAVED TS MOD&gt;"</span><span class="p">);</span>
     <span class="p">}</span>
     <span class="k">catch</span> <span class="p">(</span><span class="k">const</span> <span class="n">c10</span><span class="o">::</span><span class="n">Error</span><span class="o">&amp;</span> <span class="n">e</span><span class="p">)</span> <span class="p">{</span>
         <span class="n">std</span><span class="o">::</span><span class="n">cerr</span> <span class="o">&lt;&lt;</span> <span class="s">"error loading the model</span><span class="se">\n</span><span class="s">"</span><span class="p">;</span>

docs/tutorials/getting_started.html

@@ -84,6 +84,19 @@ all attributes accesses are replaced with explicit inputs of the graph
 graph are the trainable parameters used in this method. The remaining inputs
 are the true inputs to the function.
 
+Lower Tuples
+***************************************
+
+    `torch/csrc/jit/passes/lower_tuples.h <https://github.com/pytorch/pytorch/blob/master/torch/csrc/jit/passes/lower_tuples.h>`_
+
+* ``LowerSimpleTuples``:
+
+Removes tuples where TupleConstruct and TupleUnpack are matched but leaves tuples in place across if statements, loops, and as inputs/outputs
+
+* ``LowerAllTuples``:
+
+Removes _all_ tuples and raises an error if some cannot be removed, this is used by ONNX to ensure there are not tuples before conversion, but will not work on graphs whose inputs contain tuples.
+
 Remove Dropout
 ***************************************