public class nvinfer_plugin extends nvinfer_plugin
Modifier and Type | Class and Description |
---|---|
static class |
nvinfer_plugin.CodeTypeSSD
\enum CodeTypeSSD
\brief The type of encoding used for decoding the bounding boxes and loc_data.
|
Constructor and Description |
---|
nvinfer_plugin() |
Modifier and Type | Method and Description |
---|---|
static IPluginV2 |
createAnchorGeneratorPlugin(GridAnchorParameters param,
int numLayers)
Deprecated.
Deprecated in TensorRT 8.5. Use GridAnchorPluginCreator::createPlugin() to create an instance of
"GridAnchor_TRT" version 1 plugin.
|
static IPluginV2 |
createBatchedNMSPlugin(NMSParameters param)
Deprecated.
Deprecated in TensorRT 8.5. Use BatchedNMSPluginCreator::createPlugin() to create an instance of
"BatchedNMS_TRT" version 1 plugin.
|
static IPluginV2 |
createInstanceNormalizationPlugin(float epsilon,
Weights scale_weights,
Weights bias_weights)
Deprecated.
Deprecated in TensorRT 8.5. Use InstanceNormalizationPluginCreator::createPlugin() to create an
instance of "InstanceNormalization_TRT" version 1 plugin.
|
static IPluginV2 |
createNMSPlugin(DetectionOutputParameters param)
Deprecated.
Deprecated in TensorRT 8.5. Use NMSPluginCreator::createPlugin() to create an instance of "NMS_TRT"
version 1 plugin.
|
static IPluginV2 |
createNormalizePlugin(Weights scales,
boolean acrossSpatial,
boolean channelShared,
float eps)
Deprecated.
Deprecated in TensorRT 8.5. Use NormalizePluginCreator::createPlugin() to create an instance of
"Normalize_TRT" version 1 plugin.
|
static IPluginV2 |
createPriorBoxPlugin(PriorBoxParameters param)
Deprecated.
Deprecated in TensorRT 8.5. Use PriorBoxPluginCreator::createPlugin() to create an instance of
"PriorBox_TRT" version 1 plugin.
|
static IPluginV2 |
createRegionPlugin(RegionParameters params)
Deprecated.
Deprecated in TensorRT 8.5. Use RegionPluginCreator::createPlugin() to create an instance of
"Region_TRT" version 1 plugin.
|
static IPluginV2 |
createReorgPlugin(int stride)
Deprecated.
Deprecated in TensorRT 8.5. Use ReorgPluginCreator::createPlugin() to create an instance of
"Reorg_TRT" version 1 plugin.
|
static IPluginV2 |
createRPNROIPlugin(int featureStride,
int preNmsTop,
int nmsMaxOut,
float iouThreshold,
float minBoxSize,
float spatialScale,
DimsHW pooling,
Weights anchorRatios,
Weights anchorScales)
Deprecated.
Deprecated in TensorRT 8.5. Use RPROIPluginCreator::createPlugin() to create an instance of
"RPROI_TRT" version 1 plugin.
|
static IPluginV2 |
createSplitPlugin(int axis,
int[] output_lengths,
int noutput)
Deprecated.
|
static IPluginV2 |
createSplitPlugin(int axis,
IntBuffer output_lengths,
int noutput)
Deprecated.
|
static IPluginV2 |
createSplitPlugin(int axis,
IntPointer output_lengths,
int noutput)
Deprecated.
Deprecated in TensorRT 8.5 along with the "Split" plugin. Use INetworkDefinition::addSlice() to add
slice layer(s) as necessary to accomplish the required effect.
|
static boolean |
initLibNvInferPlugins(Pointer logger,
BytePointer libNamespace) |
static boolean |
initLibNvInferPlugins(Pointer logger,
String libNamespace)
\brief Initialize and register all the existing TensorRT plugins to the Plugin Registry with an optional
namespace.
|
map
@Deprecated public static IPluginV2 createRPNROIPlugin(int featureStride, int preNmsTop, int nmsMaxOut, float iouThreshold, float minBoxSize, float spatialScale, @ByVal DimsHW pooling, @ByVal Weights anchorRatios, @ByVal Weights anchorScales)
featureStride
- Feature stride.preNmsTop
- Number of proposals to keep before applying NMS.nmsMaxOut
- Number of remaining proposals after applying NMS.iouThreshold
- IoU threshold.minBoxSize
- Minimum allowed bounding box size before scaling.spatialScale
- Spatial scale between the input image and the last feature map.pooling
- Spatial dimensions of pooled ROIs.anchorRatios
- Aspect ratios for generating anchor windows.anchorScales
- Scales for generating anchor windows.@Deprecated public static IPluginV2 createNormalizePlugin(@Const Weights scales, @Cast(value="bool") boolean acrossSpatial, @Cast(value="bool") boolean channelShared, float eps)
scales
- Scale weights that are applied to the output tensor.acrossSpatial
- Whether to compute the norm over adjacent channels (acrossSpatial is true) or nearby
spatial locations (within channel in which case acrossSpatial is false).channelShared
- Whether the scale weight(s) is shared across channels.eps
- Epsilon for not dividing by zero.@Deprecated public static IPluginV2 createPriorBoxPlugin(@ByVal PriorBoxParameters param)
@Deprecated public static IPluginV2 createAnchorGeneratorPlugin(GridAnchorParameters param, int numLayers)
@Deprecated public static IPluginV2 createNMSPlugin(@ByVal DetectionOutputParameters param)
@Deprecated public static IPluginV2 createReorgPlugin(int stride)
stride
- Strides in H and W, it should divide both H and W. Also stride * stride should be less than or
equal to C.@Deprecated public static IPluginV2 createRegionPlugin(@ByVal RegionParameters params)
@Deprecated public static IPluginV2 createBatchedNMSPlugin(@ByVal NMSParameters param)
plugin->getWorkspaceSize(batchSize)
.@Deprecated public static IPluginV2 createSplitPlugin(int axis, IntPointer output_lengths, int noutput)
axis
- The axis to split on.output_lengths
- The lengths of the output tensors.noutput
- The number of output tensors.@Deprecated public static IPluginV2 createSplitPlugin(int axis, IntBuffer output_lengths, int noutput)
@Deprecated public static IPluginV2 createSplitPlugin(int axis, int[] output_lengths, int noutput)
@Deprecated public static IPluginV2 createInstanceNormalizationPlugin(float epsilon, @ByVal Weights scale_weights, @ByVal Weights bias_weights)
epsilon
- The epsilon value to use to avoid division by zero.scale_weights
- The input 1-dimensional scale weights of size C to scale.bias_weights
- The input 1-dimensional bias weights of size C to offset.@Cast(value="bool") public static boolean initLibNvInferPlugins(Pointer logger, String libNamespace)
logger
- Logger object to print plugin registration informationlibNamespace
- Namespace used to register all the plugins in this libraryCopyright © 2024. All rights reserved.