TensorRT 11.3.0
NvInferPythonPlugin.h File Reference
#include "NvInfer.h"

Go to the source code of this file.

Classes

class  nvinfer1::ISymExpr
 Generic interface for a scalar symbolic expression implementable by a Python plugin / TensorRT Python backend. More...
 
class  nvinfer1::ISymExprsImpl
 Impl class for ISymExprs. More...
 
class  nvinfer1::ISymExprs
 Allows for a sequence of symbolic expressions to be communicated to the TensorRT backend. More...
 
class  nvinfer1::IKernelLaunchParamsImpl
 Impl class for IKernelLaunchParams. More...
 
class  nvinfer1::IKernelLaunchParams
 Allows for kernel launch parameters to be communicated to the TensorRT backend. More...
 
class  nvinfer1::v_1_0::IPluginV3QuickCore
 
class  nvinfer1::v_1_0::IPluginV3QuickBuild
 
class  nvinfer1::v_1_0::IPluginV3QuickAOTBuild
 
class  nvinfer1::v_1_0::IPluginV3QuickRuntime
 
class  nvinfer1::v_1_0::IPluginCreatorV3Quick
 

Namespaces

namespace  nvinfer1
 The TensorRT API version 1 namespace.
 
namespace  nvinfer1::v_1_0
 Forward declare IErrorRecorder and ILogger for use in other interfaces.
 

Typedefs

using nvinfer1::IPluginV3QuickCore = v_1_0::IPluginV3QuickCore
 
using nvinfer1::IPluginV3QuickBuild = v_1_0::IPluginV3QuickBuild
 
using nvinfer1::IPluginV3QuickAOTBuild = v_1_0::IPluginV3QuickAOTBuild
 
using nvinfer1::IPluginV3QuickRuntime = v_1_0::IPluginV3QuickRuntime
 
using nvinfer1::IPluginCreatorV3Quick = v_1_0::IPluginCreatorV3Quick
 

Enumerations

enum class  nvinfer1::PluginArgType : int32_t { nvinfer1::kINT = 0 }
 Numeric type of an extra kernel input argument in an AOT Python plugin. More...
 
enum class  nvinfer1::PluginArgDataType : int32_t { nvinfer1::kINT8 = 0 , nvinfer1::kINT16 = 1 , nvinfer1::kINT32 = 2 }
 Data type of an extra kernel input argument in an AOT Python plugin. More...
 
enum class  nvinfer1::QuickPluginCreationRequest : int32_t {
  nvinfer1::kUNKNOWN = 0 , nvinfer1::kPREFER_JIT = 1 , nvinfer1::kPREFER_AOT = 2 , nvinfer1::kSTRICT_JIT = 3 ,
  nvinfer1::kSTRICT_AOT = 4
}
 Communicates preference when a quickly deployable plugin is to be added to the network. More...
 

Detailed Description

This file contains definitions for supporting the tensorrt.plugin Python module

Warning
None of the defintions here are part of the TensorRT C++ API and may not follow semantic versioning rules. TensorRT clients must not utilize them directly.

  Copyright © 2024 NVIDIA Corporation
  Privacy Policy | Manage My Privacy | Do Not Sell or Share My Data | Terms of Service | Accessibility | Corporate Policies | Product Security | Contact