diff --git a/docsrc/contributors/conversion.rst b/docsrc/contributors/conversion.rst index deb6d85a49..f19fc5eba8 100644 --- a/docsrc/contributors/conversion.rst +++ b/docsrc/contributors/conversion.rst @@ -19,7 +19,7 @@ inputs and assemble an array of resources to pass to the converter. Inputs can b * The input is an output of a node that has already been converted - * In this case the ITensor of the output has added to the to the ``value_tensor_map``, + * In this case the ITensor of the output has added to the ``value_tensor_map``, The conversion stage will add the ITensor to the list of args for the converter * The input is from a node that produces a static value @@ -32,6 +32,7 @@ inputs and assemble an array of resources to pass to the converter. Inputs can b static value has been evaluated * The input is from a node that has not been converted + * Torch-TensorRT will error out here Node Evaluation diff --git a/docsrc/contributors/lowering.rst b/docsrc/contributors/lowering.rst index 7208e9c4bf..38c4491295 100644 --- a/docsrc/contributors/lowering.rst +++ b/docsrc/contributors/lowering.rst @@ -134,7 +134,7 @@ Removes _all_ tuples and raises an error if some cannot be removed, this is used Module Fallback ***************** - `Torch-TensorRT/core/lowering/passes/module_fallback.cpp ` + `Torch-TensorRT/core/lowering/passes/module_fallback.cpp `_ Module fallback consists of two lowering passes that must be run as a pair. The first pass is run before freezing to place delimiters in the graph around modules that should run in PyTorch. The second pass marks nodes between these delimiters after freezing to signify they should run in PyTorch. diff --git a/docsrc/contributors/partitioning.rst b/docsrc/contributors/partitioning.rst index 0259c0b515..fb294c827b 100644 --- a/docsrc/contributors/partitioning.rst +++ b/docsrc/contributors/partitioning.rst @@ -3,6 +3,6 @@ Partitioning Phase ==================== -The phase is optional and enabled by the user. It instructs the compiler to seperate nodes into ones that should run in PyTorch and ones that should run in TensorRT. -Criteria for seperation include: Lack of a converter, operator is explicitly set to run in PyTorch by the user or the node has a flag which tells partitioning to -run in PyTorch by the module fallback passes. \ No newline at end of file +The phase is optional and enabled by the user. It instructs the compiler to separate nodes into ones that should run in PyTorch and ones that should run in TensorRT. +Criteria for separation include: Lack of a converter, operator is explicitly set to run in PyTorch by the user or the node has a flag which tells partitioning to +run in PyTorch by the module fallback passes. diff --git a/docsrc/contributors/phases.rst b/docsrc/contributors/phases.rst index b654bdc569..e99af77faa 100644 --- a/docsrc/contributors/phases.rst +++ b/docsrc/contributors/phases.rst @@ -21,10 +21,10 @@ TensorRT. Partitioning ^^^^^^^^^^^^^ -:ref:`partitioning +:ref:`partitioning` -The phase is optional and enabled by the user. It instructs the compiler to seperate nodes into ones that should run in PyTorch and ones that should run in TensorRT. -Criteria for seperation include: Lack of a converter, operator is explicitly set to run in PyTorch by the user or the node has a flag which tells partitioning to +The phase is optional and enabled by the user. It instructs the compiler to separate nodes into ones that should run in PyTorch and ones that should run in TensorRT. +Criteria for separation include: Lack of a converter, operator is explicitly set to run in PyTorch by the user or the node has a flag which tells partitioning to run in PyTorch by the module fallback passes. Conversion diff --git a/docsrc/contributors/system_overview.rst b/docsrc/contributors/system_overview.rst index fb552e0101..d16697e3a3 100644 --- a/docsrc/contributors/system_overview.rst +++ b/docsrc/contributors/system_overview.rst @@ -23,7 +23,7 @@ The repository is structured into: The C++ API is unstable and subject to change until the library matures, though most work is done under the hood in the core. The core has a couple major parts: The top level compiler interface which coordinates ingesting a module, lowering, -converting and generating a new module and returning it back to the user. The there are the three main phases of the +converting and generating a new module and returning it back to the user. There are the three main phases of the compiler, the lowering phase, the conversion phase, and the execution phase. .. include:: phases.rst