147 static constexpr int32_t kVALUE = 12;
183 mImpl->setName(name);
195 return mImpl->getName();
214 mImpl->setDimensions(dimensions);
227 return mImpl->getDimensions();
242 mImpl->setType(type);
254 return mImpl->getType();
269 return mImpl->setDynamicRange(min, max);
277 return mImpl->isNetworkInput();
285 return mImpl->isNetworkOutput();
307 mImpl->setBroadcastAcrossBatch(broadcastAcrossBatch);
323 return mImpl->getBroadcastAcrossBatch();
333 return mImpl->getLocation();
348 mImpl->setLocation(location);
358 return mImpl->dynamicRangeIsSet();
366 mImpl->resetDynamicRange();
376 return mImpl->getDynamicRangeMin();
386 return mImpl->getDynamicRangeMax();
405 mImpl->setAllowedFormats(formats);
418 return mImpl->getAllowedFormats();
453 return mImpl->isShapeTensor();
476 return mImpl->isExecutionTensor();
501 return mLayer->getType();
513 mLayer->setName(name);
524 return mLayer->getName();
532 return mLayer->getNbInputs();
545 return mLayer->getInput(index);
553 return mLayer->getNbOutputs();
564 return mLayer->getOutput(index);
581 return mLayer->setInput(index, tensor);
609 mLayer->setPrecision(dataType);
621 return mLayer->getPrecision();
633 return mLayer->precisionIsSet();
643 mLayer->resetPrecision();
681 mLayer->setOutputType(index, dataType);
695 return mLayer->getOutputType(index);
708 return mLayer->outputTypeIsSet(index);
720 return mLayer->resetOutputType(index);
725 apiv::VLayer* mLayer;
970 static constexpr int32_t kVALUE = 6;
1000 mImpl->setKernelSize(kernelSize);
1012 return mImpl->getKernelSize();
1024 mImpl->setNbOutputMaps(nbOutputMaps);
1034 return mImpl->getNbOutputMaps();
1050 mImpl->setStride(stride);
1060 return mImpl->getStride();
1080 return mImpl->setPadding(padding);
1092 return mImpl->getPadding();
1112 mImpl->setNbGroups(nbGroups);
1122 return mImpl->getNbGroups();
1136 mImpl->setKernelWeights(weights);
1146 return mImpl->getKernelWeights();
1161 mImpl->setBiasWeights(weights);
1171 return mImpl->getBiasWeights();
1187 return mImpl->setDilation(dilation);
1199 return mImpl->getDilation();
1216 mImpl->setPrePadding(padding);
1226 return mImpl->getPrePadding();
1243 mImpl->setPostPadding(padding);
1253 return mImpl->getPostPadding();
1267 mImpl->setPaddingMode(paddingMode);
1279 return mImpl->getPaddingMode();
1292 mImpl->setKernelSizeNd(kernelSize);
1302 return mImpl->getKernelSizeNd();
1317 mImpl->setStrideNd(stride);
1327 return mImpl->getStrideNd();
1345 mImpl->setPaddingNd(padding);
1357 return mImpl->getPaddingNd();
1371 mImpl->setDilationNd(dilation);
1381 return mImpl->getDilationNd();
1455 mImpl->setNbOutputChannels(nbOutputs);
1465 return mImpl->getNbOutputChannels();
1475 mImpl->setKernelWeights(weights);
1485 return mImpl->getKernelWeights();
1497 mImpl->setBiasWeights(weights);
1507 return mImpl->getBiasWeights();
1563 mImpl->setActivationType(type);
1573 return mImpl->getActivationType();
1588 mImpl->setAlpha(alpha);
1602 mImpl->setBeta(beta);
1611 return mImpl->getAlpha();
1620 return mImpl->getBeta();
1650 static constexpr int32_t kVALUE = 3;
1677 mImpl->setPoolingType(type);
1687 return mImpl->getPoolingType();
1701 mImpl->setWindowSize(windowSize);
1713 return mImpl->getWindowSize();
1729 mImpl->setStride(stride);
1741 return mImpl->getStride();
1757 mImpl->setPadding(padding);
1771 return mImpl->getPadding();
1786 mImpl->setBlendFactor(blendFactor);
1799 return mImpl->getBlendFactor();
1816 mImpl->setAverageCountExcludesPadding(exclusive);
1827 return mImpl->getAverageCountExcludesPadding();
1845 mImpl->setPrePadding(padding);
1855 return mImpl->getPrePadding();
1873 mImpl->setPostPadding(padding);
1883 return mImpl->getPostPadding();
1896 mImpl->setPaddingMode(paddingMode);
1907 return mImpl->getPaddingMode();
1920 mImpl->setWindowSizeNd(windowSize);
1930 return mImpl->getWindowSizeNd();
1945 mImpl->setStrideNd(stride);
1955 return mImpl->getStrideNd();
1974 mImpl->setPaddingNd(padding);
1986 return mImpl->getPaddingNd();
2017 mImpl->setWindowSize(windowSize);
2027 return mImpl->getWindowSize();
2038 mImpl->setAlpha(alpha);
2048 return mImpl->getAlpha();
2059 mImpl->setBeta(beta);
2069 return mImpl->getBeta();
2090 return mImpl->getK();
2157 mImpl->setMode(mode);
2167 return mImpl->getMode();
2177 mImpl->setShift(shift);
2187 return mImpl->getShift();
2197 mImpl->setScale(scale);
2207 return mImpl->getScale();
2217 mImpl->setPower(power);
2227 return mImpl->getPower();
2242 return mImpl->getChannelAxis();
2263 mImpl->setChannelAxis(channelAxis);
2327 mImpl->setAxes(axes);
2337 return mImpl->getAxes();
2372 mImpl->setAxis(axis);
2382 return mImpl->getAxis();
2413 mImpl->setKernelSize(kernelSize);
2425 return mImpl->getKernelSize();
2437 mImpl->setNbOutputMaps(nbOutputMaps);
2447 return mImpl->getNbOutputMaps();
2463 mImpl->setStride(stride);
2475 return mImpl->getStride();
2495 mImpl->setPadding(padding);
2509 return mImpl->getPadding();
2529 mImpl->setNbGroups(nbGroups);
2539 return mImpl->getNbGroups();
2553 mImpl->setKernelWeights(weights);
2563 return mImpl->getKernelWeights();
2578 mImpl->setBiasWeights(weights);
2588 return mImpl->getBiasWeights();
2606 mImpl->setPrePadding(padding);
2616 return mImpl->getPrePadding();
2634 mImpl->setPostPadding(padding);
2644 return mImpl->getPostPadding();
2658 mImpl->setPaddingMode(paddingMode);
2670 return mImpl->getPaddingMode();
2685 mImpl->setKernelSizeNd(kernelSize);
2695 return mImpl->getKernelSizeNd();
2712 mImpl->setStrideNd(stride);
2722 return mImpl->getStrideNd();
2740 mImpl->setPaddingNd(padding);
2752 return mImpl->getPaddingNd();
2786 mImpl->setDilationNd(dilation);
2796 return mImpl->getDilationNd();
2845 static constexpr int32_t kVALUE = 14;
2882 return mImpl->setOperation(op);
2894 return mImpl->getOperation();
3017 mImpl->setGatherAxis(axis);
3028 return mImpl->getGatherAxis();
3049 mImpl->setNbElementWiseDims(elementWiseDims);
3059 return mImpl->getNbElementWiseDims();
3069 mImpl->setMode(mode);
3079 return mImpl->getMode();
3286 return mImpl->getLayerCount();
3290 return mImpl->getHiddenSize();
3294 return mImpl->getMaxSeqLength();
3298 return mImpl->getDataLength();
3317 return mImpl->setSequenceLengths(seqLengths);
3329 return mImpl->getSequenceLengths();
3339 mImpl->setOperation(op);
3349 return mImpl->getOperation();
3359 mImpl->setInputMode(op);
3369 return mImpl->getInputMode();
3384 mImpl->setDirection(op);
3394 return mImpl->getDirection();
3453 mImpl->setWeightsForGate(layerIndex, gate, isW, weights);
3463 return mImpl->getWeightsForGate(layerIndex, gate, isW);
3488 mImpl->setBiasForGate(layerIndex, gate, isW, bias);
3498 return mImpl->getBiasForGate(layerIndex, gate, isW);
3515 mImpl->setHiddenState(hidden);
3525 return mImpl->getHiddenState();
3544 mImpl->setCellState(cell);
3554 return mImpl->getCellState();
3581 return mImpl->getPlugin();
3659 mImpl->setOperation(op);
3669 return mImpl->getOperation();
3732 mImpl->setOperation(op);
3742 return mImpl->getOperation();
3752 mImpl->setReduceAxes(reduceAxes);
3762 return mImpl->getReduceAxes();
3772 mImpl->setKeepDimensions(keepDimensions);
3782 return mImpl->getKeepDimensions();
3814 mImpl->setPrePadding(padding);
3826 return mImpl->getPrePadding();
3840 mImpl->setPostPadding(padding);
3852 return mImpl->getPostPadding();
3866 mImpl->setPrePaddingNd(padding);
3878 return mImpl->getPrePaddingNd();
3892 mImpl->setPostPaddingNd(padding);
3904 return mImpl->getPostPaddingNd();
3949 mImpl->setFirstTranspose(permutation);
3961 return mImpl->getFirstTranspose();
3986 mImpl->setReshapeDimensions(dimensions);
3999 return mImpl->getReshapeDimensions();
4046 mImpl->setSecondTranspose(permutation);
4058 return mImpl->getSecondTranspose();
4074 return mImpl->setZeroIsPlaceholder(zeroIsPlaceholder);
4087 return mImpl->getZeroIsPlaceholder();
4178 mImpl->setStart(start);
4193 return mImpl->getStart();
4207 return mImpl->setSize(size);
4222 return mImpl->getSize();
4236 mImpl->setStride(stride);
4251 return mImpl->getStride();
4261 mImpl->setMode(mode);
4271 return mImpl->getMode();
4360 mImpl->setOperation(op);
4370 return mImpl->getOperation();
4392 return mImpl->getK();
4402 mImpl->setReduceAxes(reduceAxes);
4412 return mImpl->getReduceAxes();
4496 mImpl->setOperation(index, op);
4508 return mImpl->getOperation(index);
4589 mImpl->setWeights(weights);
4599 return mImpl->getWeights();
4611 mImpl->setDimensions(dimensions);
4623 return mImpl->getDimensions();
4668 static constexpr int32_t kVALUE = 2;
4722 static constexpr int32_t kVALUE = 3;
4752 static constexpr int32_t kVALUE = 2;
4788 static constexpr int32_t kVALUE = 4;
4851 return mImpl->setOutputDimensions(dimensions);
4861 return mImpl->getOutputDimensions();
4889 void setScales(
float const* scales, int32_t nbScales)
noexcept
4891 mImpl->setScales(scales, nbScales);
4908 int32_t
getScales(int32_t size,
float* scales)
const noexcept
4910 return mImpl->getScales(size, scales);
4922 mImpl->setResizeMode(resizeMode);
4932 return mImpl->getResizeMode();
4948 mImpl->setAlignCorners(alignCorners);
4960 return mImpl->getAlignCorners();
4995 mImpl->setCoordinateTransformation(coordTransform);
5005 return mImpl->getCoordinateTransformation();
5020 mImpl->setSelectorForSinglePixel(selector);
5030 return mImpl->getSelectorForSinglePixel();
5044 mImpl->setNearestRounding(value);
5054 return mImpl->getNearestRounding();
5113 return mBoundary->getLoop();
5118 apiv::VLoopBoundaryLayer* mBoundary;
5132 return mBoundary->getConditional();
5137 apiv::VConditionalBoundaryLayer* mBoundary;
5210 return mImpl->setCondition(condition);
5226 return mImpl->addOutput(trueSubgraphOutput, falseSubgraphOutput);
5238 return mImpl->addInput(input);
5251 mImpl->setName(name);
5261 return mImpl->getName();
5320 return mImpl->getLoopOutput();
5337 mImpl->setAxis(axis);
5343 return mImpl->getAxis();
5378 return mImpl->getTripLimit();
5392 mImpl->setAxis(axis);
5398 return mImpl->getAxis();
5408 mImpl->setReverse(reverse);
5414 return mImpl->getReverse();
5438 return mImpl->addRecurrence(initialValue);
5459 return mImpl->addTripLimit(tensor, limit);
5472 return mImpl->addIterator(tensor, axis, reverse);
5484 return mImpl->addLoopOutput(tensor, outputKind, axis);
5497 mImpl->setName(name);
5507 return mImpl->getName();
5552 mImpl->setMessage(message);
5562 return mImpl->getMessage();
5634 mImpl->setDimensions(dimensions);
5649 return mImpl->getDimensions();
5659 mImpl->setOperation(op);
5669 return mImpl->getOperation();
5687 mImpl->setAlpha(alpha);
5702 return mImpl->getAlpha();
5720 mImpl->setBeta(beta);
5735 return mImpl->getBeta();
5841 return mImpl->getAxis();
5852 mImpl->setAxis(axis);
5928 return mImpl->getAxis();
5939 mImpl->setAxis(axis);
5996 return mImpl->setEquation(equation);
6006 return mImpl->getEquation();
6102 mImpl->setMode(mode);
6112 return mImpl->getMode();
6122 mImpl->setAxis(axis);
6130 return mImpl->getAxis();
6199 return mImpl->addInput(name, type, dimensions);
6213 mImpl->markOutput(tensor);
6237 return mImpl->addConvolution(input, nbOutputMaps, kernelSize, kernelWeights, biasWeights);
6260 return mImpl->addFullyConnected(input, nbOutputs, kernelWeights, biasWeights);
6279 return mImpl->addActivation(input, type);
6298 return mImpl->addPooling(input, type, windowSize);
6317 return mImpl->addLRN(input, window, alpha, beta, k);
6344 return mImpl->addScale(input, mode, shift, scale, power);
6357 return mImpl->addSoftMax(input);
6374 return mImpl->addConcatenation(inputs, nbInputs);
6398 return mImpl->addDeconvolution(input, nbOutputMaps, kernelSize, kernelWeights, biasWeights);
6425 return mImpl->addElementWise(input1, input2, op);
6447 return mImpl->addUnary(input, operation);
6464 return mImpl->addPadding(input, prePadding, postPadding);
6478 return mImpl->addShuffle(input);
6490 return mImpl->getNbLayers();
6504 return mImpl->getLayer(index);
6516 return mImpl->getNbInputs();
6532 return mImpl->getInput(index);
6546 return mImpl->getNbOutputs();
6562 return mImpl->getOutput(index);
6602 return mImpl->addReduce(input, operation, reduceAxes, keepDimensions);
6635 return mImpl->addTopK(input, op, k, reduceAxes);
6651 return mImpl->addGather(data, indices, axis);
6667 return mImpl->addGatherV2(data, indices, mode);
6685 return mImpl->addRaggedSoftMax(input, bounds);
6707 return mImpl->addMatrixMultiply(input0, op0, input1, op1);
6732 return mImpl->addConstant(dimensions, weights);
6800 ITensor& input, int32_t layerCount, int32_t hiddenSize, int32_t maxSeqLen,
RNNOperation op)
noexcept
6802 return mImpl->addRNNv2(input, layerCount, hiddenSize, maxSeqLen, op);
6816 return mImpl->addIdentity(input);
6831 mImpl->removeTensor(tensor);
6843 mImpl->unmarkOutput(tensor);
6862 return mImpl->addPluginV2(inputs, nbInputs, plugin);
6881 return mImpl->addSlice(input, start, size, stride);
6903 mImpl->setName(name);
6917 return mImpl->getName();
6935 return mImpl->addShape(input);
6953 return mImpl->hasImplicitBatchDimension();
6971 return mImpl->markOutputForShapes(tensor);
6983 return mImpl->unmarkOutputForShapes(tensor);
7001 return mImpl->addParametricReLU(input, slope);
7024 return mImpl->addConvolutionNd(input, nbOutputMaps, kernelSize, kernelWeights, biasWeights);
7043 return mImpl->addPoolingNd(input, type, windowSize);
7066 return mImpl->addDeconvolutionNd(input, nbOutputMaps, kernelSize, kernelWeights, biasWeights);
7102 return mImpl->addScaleNd(input, mode, shift, scale, power, channelAxis);
7117 return mImpl->addResize(input);
7134 return mImpl->hasExplicitPrecision();
7150 return mImpl->addLoop();
7190 return mImpl->addSelect(condition, thenInput, elseInput);
7207 return mImpl->addAssertion(condition, message);
7225 return mImpl->addFill(dimensions, op);
7242 return mImpl->addPaddingNd(input, prePadding, postPadding);
7262 return mImpl->setWeightsName(weights, name);
7281 mImpl->setErrorRecorder(recorder);
7296 return mImpl->getErrorRecorder();
7315 return mImpl->addDequantize(input, scale);
7335 return mImpl->addScatter(data, indices, updates, mode);
7354 return mImpl->addQuantize(input, scale);
7369 return mImpl->addIfConditional();
7383 return mImpl->addEinsum(inputs, nbInputs, equation);
7448 virtual
bool getBatch(
void* bindings[],
char const* names[], int32_t nbBindings) noexcept = 0;
7464 virtual
void const* readCalibrationCache(std::
size_t& length) noexcept = 0;
7474 virtual
void writeCalibrationCache(
void const* ptr, std::
size_t length) noexcept = 0;
7568 virtual
double getRegressionCutoff() const noexcept = 0;
7582 virtual
void const* readHistogramCache(std::
size_t& length) noexcept = 0;
7592 virtual
void writeHistogramCache(
void const* ptr, std::
size_t length) noexcept = 0;
7615 return mImpl->getTensorFormat();
7623 return mImpl->getDataType();
7631 return mImpl->getStrides();
7658 return mImpl->getImplementation();
7666 return mImpl->getTactic();
7691 return mImpl->getName();
7702 return mImpl->getDimensions(index, select);
7710 return mImpl->getNbInputs();
7718 return mImpl->getNbOutputs();
7750 return mImpl->getAlgorithmIOInfo(index);
7758 return mImpl->getAlgorithmVariant();
7766 return mImpl->getTimingMSec();
7774 return mImpl->getWorkspaceSize();
7787 return mImpl->getAlgorithmIOInfoByIndex(index);
7821 int32_t nbChoices, int32_t* selection)
noexcept = 0;
7833 int32_t nbAlgorithms)
noexcept = 0;
7978 return mImpl->serialize();
8002 return mImpl->combine(inputCache, ignoreMismatch);
8012 return mImpl->reset();
8097 mImpl->setMinTimingIterations(minTiming);
8111 return mImpl->getMinTimingIterations();
8124 mImpl->setAvgTimingIterations(avgTiming);
8136 return mImpl->getAvgTimingIterations();
8149 mImpl->setEngineCapability(capability);
8161 return mImpl->getEngineCapability();
8171 mImpl->setInt8Calibrator(calibrator);
8179 return mImpl->getInt8Calibrator();
8194 mImpl->setMaxWorkspaceSize(workspaceSize);
8211 return mImpl->getMaxWorkspaceSize();
8228 mImpl->setFlags(builderFlags);
8240 return mImpl->getFlags();
8252 mImpl->clearFlag(builderFlag);
8264 mImpl->setFlag(builderFlag);
8276 return mImpl->getFlag(builderFlag);
8291 mImpl->setDeviceType(layer, deviceType);
8300 return mImpl->getDeviceType(layer);
8310 return mImpl->isDeviceTypeSet(layer);
8320 mImpl->resetDeviceType(layer);
8329 return mImpl->canRunOnDLA(layer);
8344 mImpl->setDLACore(dlaCore);
8353 return mImpl->getDLACore();
8363 mImpl->setDefaultDeviceType(deviceType);
8373 return mImpl->getDefaultDeviceType();
8409 return mImpl->setProfileStream(stream);
8421 return mImpl->getProfileStream();
8437 return mImpl->addOptimizationProfile(profile);
8450 return mImpl->getNbOptimizationProfiles();
8462 mImpl->setProfilingVerbosity(verbosity);
8475 return mImpl->getProfilingVerbosity();
8484 mImpl->setAlgorithmSelector(selector);
8492 return mImpl->getAlgorithmSelector();
8507 return mImpl->setCalibrationProfile(profile);
8517 return mImpl->getCalibrationProfile();
8534 mImpl->setQuantizationFlags(flags);
8546 return mImpl->getQuantizationFlags();
8558 mImpl->clearQuantizationFlag(flag);
8570 mImpl->setQuantizationFlag(flag);
8582 return mImpl->getQuantizationFlag(flag);
8607 return mImpl->setTacticSources(tacticSources);
8622 return mImpl->getTacticSources();
8641 return mImpl->createTimingCache(blob, size);
8664 return mImpl->setTimingCache(cache, ignoreMismatch);
8674 return mImpl->getTimingCache();
8706 mImpl->setMemoryPoolLimit(pool, poolSize);
8725 return mImpl->getMemoryPoolLimit(pool);
8797 mImpl->setMaxBatchSize(batchSize);
8812 return mImpl->getMaxBatchSize();
8820 return mImpl->platformHasFastFp16();
8828 return mImpl->platformHasFastInt8();
8852 return mImpl->getMaxDLABatchSize();
8860 return mImpl->getNbDLACores();
8876 mImpl->setGpuAllocator(allocator);
8886 return mImpl->createBuilderConfig();
8902 return mImpl->buildEngineWithConfig(network, config);
8919 return mImpl->createNetworkV2(flags);
8933 return mImpl->createOptimizationProfile();
8952 mImpl->setErrorRecorder(recorder);
8967 return mImpl->getErrorRecorder();
8983 return mImpl->platformHasTf32();
9002 return mImpl->buildSerializedNetwork(network, config);
9026 return mImpl->isNetworkSupported(network, config);
9036 return mImpl->getLogger();
9050 return mImpl->setMaxThreads(maxThreads);
9064 return mImpl->getMaxThreads();
9077extern "C" TENSORRTAPI void* createInferBuilder_INTERNAL(
void* logger, int32_t version)
noexcept;
9091inline IBuilder* createInferBuilder(ILogger& logger)
noexcept
#define TENSORRTAPI
Definition: NvInferRuntimeCommon.h:54
#define NV_TENSORRT_VERSION
Definition: NvInferRuntimeCommon.h:73
#define TRT_DEPRECATED
Definition: NvInferRuntimeCommon.h:40
#define TRT_DEPRECATED_ENUM
Definition: NvInferRuntimeCommon.h:41
Definition: NvInferRuntimeCommon.h:153
static constexpr int32_t MAX_DIMS
The maximum rank (number of dimensions) supported for a tensor.
Definition: NvInferRuntimeCommon.h:156
Descriptor for two-dimensional spatial data.
Definition: NvInferLegacyDims.h:64
An Activation layer in a network definition.
Definition: NvInfer.h:1552
void setBeta(float beta) noexcept
Set the beta parameter (must be finite).
Definition: NvInfer.h:1600
void setActivationType(ActivationType type) noexcept
Set the type of activation to be performed.
Definition: NvInfer.h:1561
ActivationType getActivationType() const noexcept
Get the type of activation to be performed.
Definition: NvInfer.h:1571
float getAlpha() const noexcept
Get the alpha parameter.
Definition: NvInfer.h:1609
virtual ~IActivationLayer() noexcept=default
float getBeta() const noexcept
Get the beta parameter.
Definition: NvInfer.h:1618
void setAlpha(float alpha) noexcept
Set the alpha parameter (must be finite).
Definition: NvInfer.h:1586
Describes the context and requirements, that could be fulfilled by one or more instances of IAlgorith...
Definition: NvInfer.h:7683
int32_t getNbOutputs() const noexcept
Return number of outputs of the algorithm.
Definition: NvInfer.h:7716
int32_t getNbInputs() const noexcept
Return number of inputs of the algorithm.
Definition: NvInfer.h:7708
char const * getName() const noexcept
Return name of the algorithm node. This is a unique identifier for the IAlgorithmContext.
Definition: NvInfer.h:7689
virtual ~IAlgorithmContext() noexcept=default
Dims getDimensions(int32_t index, OptProfileSelector select) const noexcept
Get the minimum / optimum / maximum dimensions for input or output tensor.
Definition: NvInfer.h:7700
Describes a variation of execution of a layer. An algorithm is represented by IAlgorithmVariant and t...
Definition: NvInfer.h:7736
std::size_t getWorkspaceSize() const noexcept
The size of the GPU temporary memory in bytes which the algorithm uses at execution time.
Definition: NvInfer.h:7772
float getTimingMSec() const noexcept
The time in milliseconds to execute the algorithm.
Definition: NvInfer.h:7764
IAlgorithmIOInfo const * getAlgorithmIOInfoByIndex(int32_t index) const noexcept
Returns the format of an Algorithm input or output. Algorithm inputs are incrementally numbered first...
Definition: NvInfer.h:7785
virtual ~IAlgorithm() noexcept=default
TRT_DEPRECATED IAlgorithmIOInfo const & getAlgorithmIOInfo(int32_t index) const noexcept
Returns the format of an Algorithm input or output. Algorithm inputs are incrementally numbered first...
Definition: NvInfer.h:7748
IAlgorithmVariant const & getAlgorithmVariant() const noexcept
Returns the algorithm variant.
Definition: NvInfer.h:7756
Carries information about input or output of the algorithm. IAlgorithmIOInfo for all the input and ou...
Definition: NvInfer.h:7608
virtual ~IAlgorithmIOInfo() noexcept=default
Dims getStrides() const noexcept
Return strides of the input/output tensor of algorithm.
Definition: NvInfer.h:7629
DataType getDataType() const noexcept
Return DataType of the input/output of algorithm.
Definition: NvInfer.h:7621
TensorFormat getTensorFormat() const noexcept
Return TensorFormat of the input/output of algorithm.
Definition: NvInfer.h:7613
Interface implemented by application for selecting and reporting algorithms of a layer provided by th...
Definition: NvInfer.h:7804
virtual void reportAlgorithms(IAlgorithmContext const *const *algoContexts, IAlgorithm const *const *algoChoices, int32_t nbAlgorithms) noexcept=0
Called by TensorRT to report choices it made.
virtual int32_t selectAlgorithms(IAlgorithmContext const &context, IAlgorithm const *const *choices, int32_t nbChoices, int32_t *selection) noexcept=0
Select Algorithms for a layer from the given list of algorithm choices.
virtual ~IAlgorithmSelector() noexcept=default
provides a unique 128-bit identifier, which along with the input and output information denotes the v...
Definition: NvInfer.h:7651
virtual ~IAlgorithmVariant() noexcept=default
int64_t getTactic() const noexcept
Return tactic of the algorithm.
Definition: NvInfer.h:7664
int64_t getImplementation() const noexcept
Return implementation of the algorithm.
Definition: NvInfer.h:7656
An assertion layer in a network.
Definition: NvInfer.h:5540
void setMessage(char const *message) noexcept
Set the message to print if the assertion fails.
Definition: NvInfer.h:5550
char const * getMessage() const noexcept
Return the assertion message.
Definition: NvInfer.h:5560
virtual ~IAssertionLayer() noexcept=default
Holds properties for configuring a builder to produce an engine.
Definition: NvInfer.h:8079
virtual TRT_DEPRECATED int32_t getMinTimingIterations() const noexcept
Query the number of minimization iterations.
Definition: NvInfer.h:8109
IOptimizationProfile const * getCalibrationProfile() noexcept
Get the current calibration profile.
Definition: NvInfer.h:8515
void setMemoryPoolLimit(MemoryPoolType pool, std::size_t poolSize) noexcept
Set the memory size for the memory pool.
Definition: NvInfer.h:8704
void setQuantizationFlag(QuantizationFlag flag) noexcept
Set a single quantization flag.
Definition: NvInfer.h:8568
nvinfer1::ITimingCache * createTimingCache(void const *blob, std::size_t size) const noexcept
Create timing cache.
Definition: NvInfer.h:8639
void setInt8Calibrator(IInt8Calibrator *calibrator) noexcept
Set Int8 Calibration interface.
Definition: NvInfer.h:8169
void clearQuantizationFlag(QuantizationFlag flag) noexcept
clear a quantization flag.
Definition: NvInfer.h:8556
bool setTacticSources(TacticSources tacticSources) noexcept
Set tactic sources.
Definition: NvInfer.h:8605
bool getQuantizationFlag(QuantizationFlag flag) const noexcept
Returns true if the quantization flag is set.
Definition: NvInfer.h:8580
std::size_t getMemoryPoolLimit(MemoryPoolType pool) const noexcept
Get the memory size limit of the memory pool.
Definition: NvInfer.h:8723
int32_t getDLACore() const noexcept
Get the DLA core that the engine executes on.
Definition: NvInfer.h:8351
void setDeviceType(ILayer const *layer, DeviceType deviceType) noexcept
Set the device that this layer must execute on.
Definition: NvInfer.h:8289
void setEngineCapability(EngineCapability capability) noexcept
Configure the builder to target specified EngineCapability flow.
Definition: NvInfer.h:8147
bool getFlag(BuilderFlag builderFlag) const noexcept
Returns true if the build mode flag is set.
Definition: NvInfer.h:8274
void setQuantizationFlags(QuantizationFlags flags) noexcept
Set the quantization flags.
Definition: NvInfer.h:8532
bool setCalibrationProfile(IOptimizationProfile const *profile) noexcept
Add a calibration profile.
Definition: NvInfer.h:8505
virtual void setAvgTimingIterations(int32_t avgTiming) noexcept
Set the number of averaging iterations used when timing layers.
Definition: NvInfer.h:8122
void setProfilingVerbosity(ProfilingVerbosity verbosity) noexcept
Set verbosity level of layer information exposed in NVTX annotations and IEngineInspector.
Definition: NvInfer.h:8460
IAlgorithmSelector * getAlgorithmSelector() const noexcept
Get Algorithm Selector.
Definition: NvInfer.h:8490
int32_t getNbOptimizationProfiles() const noexcept
Get number of optimization profiles.
Definition: NvInfer.h:8448
QuantizationFlags getQuantizationFlags() const noexcept
Get the quantization flags.
Definition: NvInfer.h:8544
nvinfer1::ITimingCache const * getTimingCache() const noexcept
Get the pointer to the timing cache from current IBuilderConfig.
Definition: NvInfer.h:8672
void reset() noexcept
Resets the builder configuration to defaults.
Definition: NvInfer.h:8381
bool setTimingCache(ITimingCache const &cache, bool ignoreMismatch) noexcept
Attach a timing cache to IBuilderConfig.
Definition: NvInfer.h:8662
TRT_DEPRECATED void destroy() noexcept
Delete this IBuilderConfig.
Definition: NvInfer.h:8395
EngineCapability getEngineCapability() const noexcept
Query EngineCapability flow configured for the builder.
Definition: NvInfer.h:8159
void setAlgorithmSelector(IAlgorithmSelector *selector) noexcept
Set Algorithm Selector.
Definition: NvInfer.h:8482
TRT_DEPRECATED void setMaxWorkspaceSize(std::size_t workspaceSize) noexcept
Set the maximum workspace size.
Definition: NvInfer.h:8192
TRT_DEPRECATED std::size_t getMaxWorkspaceSize() const noexcept
Get the maximum workspace size.
Definition: NvInfer.h:8209
DeviceType getDefaultDeviceType() const noexcept
Get the default DeviceType which was set by setDefaultDeviceType.
Definition: NvInfer.h:8371
BuilderFlags getFlags() const noexcept
Get the build mode flags for this builder config. Defaults to 0.
Definition: NvInfer.h:8238
void setFlags(BuilderFlags builderFlags) noexcept
Set the build mode flags to turn on builder options for this network.
Definition: NvInfer.h:8226
TacticSources getTacticSources() const noexcept
Get tactic sources.
Definition: NvInfer.h:8620
void resetDeviceType(ILayer const *layer) noexcept
reset the DeviceType for this layer
Definition: NvInfer.h:8318
void setDLACore(int32_t dlaCore) noexcept
Sets the DLA core used by the network. Defaults to -1.
Definition: NvInfer.h:8342
void clearFlag(BuilderFlag builderFlag) noexcept
clear a single build mode flag.
Definition: NvInfer.h:8250
int32_t addOptimizationProfile(IOptimizationProfile const *profile) noexcept
Add an optimization profile.
Definition: NvInfer.h:8435
apiv::VBuilderConfig * mImpl
Definition: NvInfer.h:8729
int32_t getAvgTimingIterations() const noexcept
Query the number of averaging iterations.
Definition: NvInfer.h:8134
void setDefaultDeviceType(DeviceType deviceType) noexcept
Sets the default DeviceType to be used by the builder. It ensures that all the layers that can run on...
Definition: NvInfer.h:8361
void setFlag(BuilderFlag builderFlag) noexcept
Set a single build mode flag.
Definition: NvInfer.h:8262
virtual ~IBuilderConfig() noexcept=default
DeviceType getDeviceType(ILayer const *layer) const noexcept
Get the device that this layer executes on.
Definition: NvInfer.h:8298
bool canRunOnDLA(ILayer const *layer) const noexcept
Checks if a layer can run on DLA.
Definition: NvInfer.h:8327
cudaStream_t getProfileStream() const noexcept
Get the cuda stream that is used to profile this network.
Definition: NvInfer.h:8419
IInt8Calibrator * getInt8Calibrator() const noexcept
Get Int8 Calibration interface.
Definition: NvInfer.h:8177
ProfilingVerbosity getProfilingVerbosity() const noexcept
Get verbosity level of layer information exposed in NVTX annotations and IEngineInspector.
Definition: NvInfer.h:8473
bool isDeviceTypeSet(ILayer const *layer) const noexcept
whether the DeviceType has been explicitly set for this layer
Definition: NvInfer.h:8308
void setProfileStream(const cudaStream_t stream) noexcept
Set the cuda stream that is used to profile this network.
Definition: NvInfer.h:8407
Builds an engine from a network definition.
Definition: NvInfer.h:8781
int32_t getMaxDLABatchSize() const noexcept
Get the maximum batch size DLA can support. For any tensor the total volume of index dimensions combi...
Definition: NvInfer.h:8850
int32_t getNbDLACores() const noexcept
Return the number of DLA engines available to this builder.
Definition: NvInfer.h:8858
IErrorRecorder * getErrorRecorder() const noexcept
get the ErrorRecorder assigned to this interface.
Definition: NvInfer.h:8965
apiv::VBuilder * mImpl
Definition: NvInfer.h:9068
ILogger * getLogger() const noexcept
get the logger with which the builder was created
Definition: NvInfer.h:9034
bool isNetworkSupported(INetworkDefinition const &network, IBuilderConfig const &config) const noexcept
Checks that a network is within the scope of the IBuilderConfig settings.
Definition: NvInfer.h:9024
bool platformHasTf32() const noexcept
Determine whether the platform has TF32 support.
Definition: NvInfer.h:8981
int32_t getMaxThreads() const noexcept
get the maximum number of threads that can be used by the builder.
Definition: NvInfer.h:9062
TRT_DEPRECATED void destroy() noexcept
Destroy this object.
Definition: NvInfer.h:8838
nvinfer1::IOptimizationProfile * createOptimizationProfile() noexcept
Create a new optimization profile.
Definition: NvInfer.h:8931
bool platformHasFastFp16() const noexcept
Determine whether the platform has fast native fp16.
Definition: NvInfer.h:8818
void setGpuAllocator(IGpuAllocator *allocator) noexcept
Set the GPU allocator.
Definition: NvInfer.h:8874
nvinfer1::INetworkDefinition * createNetworkV2(NetworkDefinitionCreationFlags flags) noexcept
Create a network definition object.
Definition: NvInfer.h:8917
nvinfer1::IBuilderConfig * createBuilderConfig() noexcept
Create a builder configuration object.
Definition: NvInfer.h:8884
void reset() noexcept
Resets the builder state to default values.
Definition: NvInfer.h:8973
bool setMaxThreads(int32_t maxThreads) noexcept
Set the maximum number of threads.
Definition: NvInfer.h:9048
void setErrorRecorder(IErrorRecorder *recorder) noexcept
Set the ErrorRecorder for this interface.
Definition: NvInfer.h:8950
bool platformHasFastInt8() const noexcept
Determine whether the platform has fast native int8.
Definition: NvInfer.h:8826
nvinfer1::IHostMemory * buildSerializedNetwork(INetworkDefinition &network, IBuilderConfig &config) noexcept
Builds and serializes a network for the given INetworkDefinition and IBuilderConfig.
Definition: NvInfer.h:9000
virtual ~IBuilder() noexcept=default
TRT_DEPRECATED int32_t getMaxBatchSize() const noexcept
Get the maximum batch size.
Definition: NvInfer.h:8810
TRT_DEPRECATED nvinfer1::ICudaEngine * buildEngineWithConfig(INetworkDefinition &network, IBuilderConfig &config) noexcept
Builds an engine for the given INetworkDefinition and given IBuilderConfig.
Definition: NvInfer.h:8899
A concatenation layer in a network definition.
Definition: NvInfer.h:2358
void setAxis(int32_t axis) noexcept
Set the axis along which concatenation occurs.
Definition: NvInfer.h:2370
int32_t getAxis() const noexcept
Get the axis along which concatenation occurs.
Definition: NvInfer.h:2380
virtual ~IConcatenationLayer() noexcept=default
Definition: NvInfer.h:5144
virtual ~IConditionLayer() noexcept=default
Layer that represents a constant value.
Definition: NvInfer.h:4576
void setWeights(Weights weights) noexcept
Set the weights for the layer.
Definition: NvInfer.h:4587
Weights getWeights() const noexcept
Get the weights for the layer.
Definition: NvInfer.h:4597
apiv::VConstantLayer * mImpl
Definition: NvInfer.h:4627
void setDimensions(Dims dimensions) noexcept
Set the dimensions for the layer.
Definition: NvInfer.h:4609
virtual ~IConstantLayer() noexcept=default
Dims getDimensions() const noexcept
Get the dimensions for the layer.
Definition: NvInfer.h:4621
A convolution layer in a network definition.
Definition: NvInfer.h:987
TRT_DEPRECATED DimsHW getPadding() const noexcept
Get the padding of the convolution. If the padding is asymmetric, the pre-padding is returned.
Definition: NvInfer.h:1090
TRT_DEPRECATED DimsHW getStride() const noexcept
Get the stride of the convolution.
Definition: NvInfer.h:1058
TRT_DEPRECATED DimsHW getDilation() const noexcept
Get the dilation for a convolution.
Definition: NvInfer.h:1197
void setStrideNd(Dims stride) noexcept
Set the multi-dimension stride of the convolution.
Definition: NvInfer.h:1315
void setKernelSizeNd(Dims kernelSize) noexcept
Set the multi-dimension kernel size of the convolution.
Definition: NvInfer.h:1290
Dims getPrePadding() const noexcept
Get the pre-padding.
Definition: NvInfer.h:1224
Weights getBiasWeights() const noexcept
Get the bias weights for the convolution.
Definition: NvInfer.h:1169
int32_t getNbGroups() const noexcept
Get the number of groups of the convolution.
Definition: NvInfer.h:1120
void setPaddingMode(PaddingMode paddingMode) noexcept
Set the padding mode.
Definition: NvInfer.h:1265
TRT_DEPRECATED void setStride(DimsHW stride) noexcept
Get the stride of the convolution.
Definition: NvInfer.h:1048
Dims getPaddingNd() const noexcept
Get the multi-dimension padding of the convolution.
Definition: NvInfer.h:1355
Dims getStrideNd() const noexcept
Get the multi-dimension stride of the convolution.
Definition: NvInfer.h:1325
TRT_DEPRECATED void setPadding(DimsHW padding) noexcept
Set the padding of the convolution.
Definition: NvInfer.h:1078
Weights getKernelWeights() const noexcept
Get the kernel weights of the convolution.
Definition: NvInfer.h:1144
void setPaddingNd(Dims padding) noexcept
Set the multi-dimension padding of the convolution.
Definition: NvInfer.h:1343
Dims getDilationNd() const noexcept
Get the multi-dimension dilation of the convolution.
Definition: NvInfer.h:1379
void setKernelWeights(Weights weights) noexcept
Set the kernel weights for the convolution.
Definition: NvInfer.h:1134
void setDilationNd(Dims dilation) noexcept
Set the multi-dimension dilation of the convolution.
Definition: NvInfer.h:1369
Dims getPostPadding() const noexcept
Get the post-padding.
Definition: NvInfer.h:1251
TRT_DEPRECATED void setDilation(DimsHW dilation) noexcept
Set the dilation for a convolution.
Definition: NvInfer.h:1185
void setNbGroups(int32_t nbGroups) noexcept
Set the number of groups for a convolution.
Definition: NvInfer.h:1110
PaddingMode getPaddingMode() const noexcept
Get the padding mode.
Definition: NvInfer.h:1277
int32_t getNbOutputMaps() const noexcept
Get the number of output maps for the convolution.
Definition: NvInfer.h:1032
void setPrePadding(Dims padding) noexcept
Set the multi-dimension pre-padding of the convolution.
Definition: NvInfer.h:1214
virtual ~IConvolutionLayer() noexcept=default
void setBiasWeights(Weights weights) noexcept
Set the bias weights for the convolution.
Definition: NvInfer.h:1159
Dims getKernelSizeNd() const noexcept
Get the multi-dimension kernel size of the convolution.
Definition: NvInfer.h:1300
void setNbOutputMaps(int32_t nbOutputMaps) noexcept
Set the number of output maps for the convolution.
Definition: NvInfer.h:1022
TRT_DEPRECATED void setKernelSize(DimsHW kernelSize) noexcept
Set the HW kernel size of the convolution.
Definition: NvInfer.h:998
void setPostPadding(Dims padding) noexcept
Set the multi-dimension post-padding of the convolution.
Definition: NvInfer.h:1241
TRT_DEPRECATED DimsHW getKernelSize() const noexcept
Get the HW kernel size of the convolution.
Definition: NvInfer.h:1010
An engine for executing inference on a built network, with functionally unsafe features.
Definition: NvInferRuntime.h:1343
A deconvolution layer in a network definition.
Definition: NvInfer.h:2398
void setBiasWeights(Weights weights) noexcept
Set the bias weights for the deconvolution.
Definition: NvInfer.h:2576
TRT_DEPRECATED void setStride(DimsHW stride) noexcept
Set the stride of the deconvolution.
Definition: NvInfer.h:2461
void setNbOutputMaps(int32_t nbOutputMaps) noexcept
Set the number of output feature maps for the deconvolution.
Definition: NvInfer.h:2435
Weights getKernelWeights() const noexcept
Get the kernel weights for the deconvolution.
Definition: NvInfer.h:2561
void setPrePadding(Dims padding) noexcept
Set the multi-dimension pre-padding of the deconvolution.
Definition: NvInfer.h:2604
TRT_DEPRECATED DimsHW getPadding() const noexcept
Get the padding of the deconvolution.
Definition: NvInfer.h:2507
Dims getStrideNd() const noexcept
Get the multi-dimension stride of the deconvolution.
Definition: NvInfer.h:2720
int32_t getNbGroups() const noexcept
Get the number of groups for a deconvolution.
Definition: NvInfer.h:2537
Dims getDilationNd() const noexcept
Get the multi-dimension dilation of the deconvolution.
Definition: NvInfer.h:2794
Weights getBiasWeights() const noexcept
Get the bias weights for the deconvolution.
Definition: NvInfer.h:2586
void setKernelWeights(Weights weights) noexcept
Set the kernel weights for the deconvolution.
Definition: NvInfer.h:2551
void setDilationNd(Dims dilation) noexcept
Set the multi-dimension dilation of the deconvolution.
Definition: NvInfer.h:2784
TRT_DEPRECATED DimsHW getStride() const noexcept
Get the stride of the deconvolution.
Definition: NvInfer.h:2473
TRT_DEPRECATED void setPadding(DimsHW padding) noexcept
Set the padding of the deconvolution.
Definition: NvInfer.h:2493
Dims getPostPadding() const noexcept
Get the padding.
Definition: NvInfer.h:2642
Dims getKernelSizeNd() const noexcept
Get the multi-dimension kernel size of the deconvolution.
Definition: NvInfer.h:2693
void setKernelSizeNd(Dims kernelSize) noexcept
Set the multi-dimension kernel size of the deconvolution.
Definition: NvInfer.h:2683
virtual ~IDeconvolutionLayer() noexcept=default
TRT_DEPRECATED void setKernelSize(DimsHW kernelSize) noexcept
Set the HW kernel size of the convolution.
Definition: NvInfer.h:2411
void setStrideNd(Dims stride) noexcept
Set the multi-dimension stride of the deconvolution.
Definition: NvInfer.h:2710
Dims getPaddingNd() const noexcept
Get the multi-dimension padding of the deconvolution.
Definition: NvInfer.h:2750
void setNbGroups(int32_t nbGroups) noexcept
Set the number of groups for a deconvolution.
Definition: NvInfer.h:2527
void setPaddingMode(PaddingMode paddingMode) noexcept
Set the padding mode.
Definition: NvInfer.h:2656
void setPaddingNd(Dims padding) noexcept
Set the multi-dimension padding of the deconvolution.
Definition: NvInfer.h:2738
TRT_DEPRECATED DimsHW getKernelSize() const noexcept
Get the HW kernel size of the deconvolution.
Definition: NvInfer.h:2423
void setPostPadding(Dims padding) noexcept
Set the multi-dimension post-padding of the deconvolution.
Definition: NvInfer.h:2632
int32_t getNbOutputMaps() const noexcept
Get the number of output feature maps for the deconvolution.
Definition: NvInfer.h:2445
Dims getPrePadding() const noexcept
Get the pre-padding.
Definition: NvInfer.h:2614
PaddingMode getPaddingMode() const noexcept
Get the padding mode.
Definition: NvInfer.h:2668
A Dequantize layer in a network definition.
Definition: NvInfer.h:5916
virtual ~IDequantizeLayer() noexcept=default
int32_t getAxis() const noexcept
Get the quantization axis.
Definition: NvInfer.h:5926
void setAxis(int32_t axis) noexcept
Set the quantization axis.
Definition: NvInfer.h:5937
An Einsum layer in a network.
Definition: NvInfer.h:5983
bool setEquation(char const *equation) noexcept
Set the equation. The equation is a comma-separated list of subscript labels, where each label refers...
Definition: NvInfer.h:5994
virtual ~IEinsumLayer() noexcept=default
char const * getEquation() const noexcept
Return the equation.
Definition: NvInfer.h:6004
A elementwise layer in a network definition.
Definition: NvInfer.h:2869
virtual ~IElementWiseLayer() noexcept=default
apiv::VElementWiseLayer * mImpl
Definition: NvInfer.h:2898
ElementWiseOperation getOperation() const noexcept
Get the binary operation for the layer.
Definition: NvInfer.h:2892
void setOperation(ElementWiseOperation op) noexcept
Set the binary operation for the layer.
Definition: NvInfer.h:2880
Reference counted application-implemented error reporting interface for TensorRT objects.
Definition: NvInferRuntimeCommon.h:1665
Generate an output tensor with specified mode.
Definition: NvInfer.h:5621
FillOperation getOperation() const noexcept
Get the fill operation for the layer.
Definition: NvInfer.h:5667
void setOperation(FillOperation op) noexcept
Set the fill operation for the layer.
Definition: NvInfer.h:5657
void setDimensions(Dims dimensions) noexcept
Set the output tensor's dimensions.
Definition: NvInfer.h:5632
void setBeta(double beta) noexcept
Set the beta parameter.
Definition: NvInfer.h:5718
double getAlpha() const noexcept
Get the value of alpha parameter.
Definition: NvInfer.h:5700
void setAlpha(double alpha) noexcept
Set the alpha parameter.
Definition: NvInfer.h:5685
Dims getDimensions() const noexcept
Get the output tensor's dimensions.
Definition: NvInfer.h:5647
double getBeta() const noexcept
Get the value of beta parameter.
Definition: NvInfer.h:5733
virtual ~IFillLayer() noexcept=default
A fully connected layer in a network definition. This layer expects an input tensor of three or more ...
Definition: NvInfer.h:1444
virtual ~IFullyConnectedLayer() noexcept=default
void setNbOutputChannels(int32_t nbOutputs) noexcept
Set the number of output channels K from the fully connected layer.
Definition: NvInfer.h:1453
Weights getKernelWeights() const noexcept
Get the kernel weights.
Definition: NvInfer.h:1483
void setBiasWeights(Weights weights) noexcept
Set the bias weights.
Definition: NvInfer.h:1495
void setKernelWeights(Weights weights) noexcept
Set the kernel weights, given as a KxC matrix in row-major order.
Definition: NvInfer.h:1473
int32_t getNbOutputChannels() const noexcept
Get the number of output channels K from the fully connected layer.
Definition: NvInfer.h:1463
Weights getBiasWeights() const noexcept
Get the bias weights.
Definition: NvInfer.h:1505
A Gather layer in a network definition. Supports several kinds of gathering.
Definition: NvInfer.h:3004
void setGatherAxis(int32_t axis) noexcept
Set the axis used by GatherMode::kELEMENTS and GatherMode::kDEFAULT The axis must be less than the nu...
Definition: NvInfer.h:3015
void setNbElementWiseDims(int32_t elementWiseDims) noexcept
Set the number of leading dimensions of indices tensor to be handled elementwise. The gathering of in...
Definition: NvInfer.h:3047
apiv::VGatherLayer * mImpl
Definition: NvInfer.h:3083
int32_t getNbElementWiseDims() const noexcept
Get the number of leading dimensions of indices tensor to be handled elementwise.
Definition: NvInfer.h:3057
void setMode(GatherMode mode) noexcept
Set the gather mode.
Definition: NvInfer.h:3067
int32_t getGatherAxis() const noexcept
Get the axis to gather on.
Definition: NvInfer.h:3026
GatherMode getMode() const noexcept
Get the gather mode.
Definition: NvInfer.h:3077
virtual ~IGatherLayer() noexcept=default
Application-implemented class for controlling allocation on the GPU.
Definition: NvInferRuntimeCommon.h:1338
Class to handle library allocated memory that is accessible to the user.
Definition: NvInferRuntime.h:144
A layer that represents the identity function.
Definition: NvInfer.h:4561
apiv::VIdentityLayer * mImpl
Definition: NvInfer.h:4563
virtual ~IIdentityLayer() noexcept=default
Definition: NvInfer.h:5127
IIfConditional * getConditional() const noexcept
Return pointer to the IIfConditional associated with this boundary layer.
Definition: NvInfer.h:5130
virtual ~IIfConditionalBoundaryLayer() noexcept=default
Definition: NvInfer.h:5197
IIfConditionalInputLayer * addInput(ITensor &input) noexcept
Add an If-conditional input.
Definition: NvInfer.h:5236
char const * getName() const noexcept
Return the name of the conditional.
Definition: NvInfer.h:5259
virtual ~IIfConditional() noexcept=default
IConditionLayer * setCondition(ITensor &condition) noexcept
Set the condition tensor for this If-Conditional construct.
Definition: NvInfer.h:5208
IIfConditionalOutputLayer * addOutput(ITensor &trueSubgraphOutput, ITensor &falseSubgraphOutput) noexcept
Add an If-conditional output.
Definition: NvInfer.h:5224
void setName(char const *name) noexcept
Set the name of the conditional.
Definition: NvInfer.h:5249
Definition: NvInfer.h:5157
virtual ~IIfConditionalOutputLayer() noexcept=default
Application-implemented interface for calibration.
Definition: NvInfer.h:7426
virtual int32_t getBatchSize() const noexcept=0
Get the batch size used for calibration batches.
Definition: NvInfer.h:7509
CalibrationAlgoType getAlgorithm() noexcept override
Definition: NvInfer.h:7514
virtual ~IInt8EntropyCalibrator2() noexcept=default
Definition: NvInfer.h:7491
virtual ~IInt8EntropyCalibrator() noexcept=default
CalibrationAlgoType getAlgorithm() noexcept override
Definition: NvInfer.h:7496
Definition: NvInfer.h:7544
CalibrationAlgoType getAlgorithm() noexcept override
Definition: NvInfer.h:7549
virtual double getQuantile() const noexcept=0
The quantile (between 0 and 1) that will be used to select the region maximum when the quantile metho...
Definition: NvInfer.h:7526
CalibrationAlgoType getAlgorithm() noexcept override
Definition: NvInfer.h:7531
virtual ~IInt8MinMaxCalibrator() noexcept=default
Definition: NvInfer.h:5387
virtual ~IIteratorLayer() noexcept=default
void setReverse(bool reverse) noexcept
Definition: NvInfer.h:5406
bool getReverse() const noexcept
True if and only if reversing input.
Definition: NvInfer.h:5412
int32_t getAxis() const noexcept
Get axis being iterated over.
Definition: NvInfer.h:5396
void setAxis(int32_t axis) noexcept
Set axis to iterate over.
Definition: NvInfer.h:5390
A LRN layer in a network definition.
Definition: NvInfer.h:2004
int32_t getWindowSize() const noexcept
Get the LRN window size.
Definition: NvInfer.h:2025
float getAlpha() const noexcept
Get the LRN alpha value.
Definition: NvInfer.h:2046
void setWindowSize(int32_t windowSize) noexcept
Set the LRN window size.
Definition: NvInfer.h:2015
void setK(float k) noexcept
Set the LRN K value.
Definition: NvInfer.h:2078
void setAlpha(float alpha) noexcept
Set the LRN alpha value.
Definition: NvInfer.h:2036
void setBeta(float beta) noexcept
Set the LRN beta value.
Definition: NvInfer.h:2057
virtual ~ILRNLayer() noexcept=default
float getBeta() const noexcept
Get the LRN beta value.
Definition: NvInfer.h:2067
float getK() const noexcept
Get the LRN K value.
Definition: NvInfer.h:2088
Base class for all layer classes in a network definition.
Definition: NvInfer.h:492
bool precisionIsSet() const noexcept
whether the computational precision has been set for this layer
Definition: NvInfer.h:631
void setPrecision(DataType dataType) noexcept
Set the computational precision of this layer.
Definition: NvInfer.h:607
void setName(char const *name) noexcept
Set the name of a layer.
Definition: NvInfer.h:511
void resetPrecision() noexcept
reset the computational precision for this layer
Definition: NvInfer.h:641
int32_t getNbInputs() const noexcept
Get the number of inputs of a layer.
Definition: NvInfer.h:530
DataType getOutputType(int32_t index) const noexcept
get the output type of this layer
Definition: NvInfer.h:693
DataType getPrecision() const noexcept
get the computational precision of this layer
Definition: NvInfer.h:619
char const * getName() const noexcept
Return the name of a layer.
Definition: NvInfer.h:522
int32_t getNbOutputs() const noexcept
Get the number of outputs of a layer.
Definition: NvInfer.h:551
bool outputTypeIsSet(int32_t index) const noexcept
whether the output type has been set for this layer
Definition: NvInfer.h:706
ITensor * getOutput(int32_t index) const noexcept
Get the layer output corresponding to the given index.
Definition: NvInfer.h:562
void setInput(int32_t index, ITensor &tensor) noexcept
Replace an input of this layer with a specific tensor.
Definition: NvInfer.h:579
void resetOutputType(int32_t index) noexcept
reset the output type for this layer
Definition: NvInfer.h:718
ITensor * getInput(int32_t index) const noexcept
Get the layer input corresponding to the given index.
Definition: NvInfer.h:543
void setOutputType(int32_t index, DataType dataType) noexcept
Set the output type of this layer.
Definition: NvInfer.h:679
LayerType getType() const noexcept
Return the type of a layer.
Definition: NvInfer.h:499
virtual ~ILayer() noexcept=default
Application-implemented logging interface for the builder, refitter and runtime.
Definition: NvInferRuntimeCommon.h:1476
Definition: NvInfer.h:5108
virtual ~ILoopBoundaryLayer() noexcept=default
ILoop * getLoop() const noexcept
Return pointer to ILoop associated with this boundary layer.
Definition: NvInfer.h:5111
Definition: NvInfer.h:5428
void setName(char const *name) noexcept
Set the name of the loop.
Definition: NvInfer.h:5495
ITripLimitLayer * addTripLimit(ITensor &tensor, TripLimit limit) noexcept
Add a trip-count limiter, based on the given tensor.
Definition: NvInfer.h:5457
IIteratorLayer * addIterator(ITensor &tensor, int32_t axis=0, bool reverse=false) noexcept
Return layer that subscripts tensor by loop iteration.
Definition: NvInfer.h:5470
ILoopOutputLayer * addLoopOutput(ITensor &tensor, LoopOutput outputKind, int32_t axis=0) noexcept
Make an output for this loop, based on the given tensor.
Definition: NvInfer.h:5482
virtual ~ILoop() noexcept=default
char const * getName() const noexcept
Return the name of the loop.
Definition: NvInfer.h:5505
IRecurrenceLayer * addRecurrence(ITensor &initialValue) noexcept
Create a recurrence layer for this loop with initialValue as its first input.
Definition: NvInfer.h:5436
Definition: NvInfer.h:5316
virtual ~ILoopOutputLayer() noexcept=default
int32_t getAxis() const noexcept
Get axis being concatenated over.
Definition: NvInfer.h:5341
LoopOutput getLoopOutput() const noexcept
Definition: NvInfer.h:5318
void setAxis(int32_t axis) noexcept
Set where to insert the contenation axis. Ignored if getLoopOutput() is kLAST_VALUE.
Definition: NvInfer.h:5335
Layer that represents a Matrix Multiplication.
Definition: NvInfer.h:4486
apiv::VMatrixMultiplyLayer * mImpl
Definition: NvInfer.h:4512
virtual ~IMatrixMultiplyLayer() noexcept=default
MatrixOperation getOperation(int32_t index) const noexcept
Get the operation for an input tensor.
Definition: NvInfer.h:4506
void setOperation(int32_t index, MatrixOperation op) noexcept
Set the operation for an input tensor.
Definition: NvInfer.h:4494
A network definition for input to the builder.
Definition: NvInfer.h:6158
IPluginV2Layer * addPluginV2(ITensor *const *inputs, int32_t nbInputs, IPluginV2 &plugin) noexcept
Add a plugin layer to the network using the IPluginV2 interface.
Definition: NvInfer.h:6860
IConvolutionLayer * addConvolutionNd(ITensor &input, int32_t nbOutputMaps, Dims kernelSize, Weights kernelWeights, Weights biasWeights) noexcept
Add a multi-dimension convolution layer to the network.
Definition: NvInfer.h:7021
IDequantizeLayer * addDequantize(ITensor &input, ITensor &scale) noexcept
Add a dequantization layer to the network.
Definition: NvInfer.h:7313
IConcatenationLayer * addConcatenation(ITensor *const *inputs, int32_t nbInputs) noexcept
Add a concatenation layer to the network.
Definition: NvInfer.h:6372
TRT_DEPRECATED IDeconvolutionLayer * addDeconvolution(ITensor &input, int32_t nbOutputMaps, DimsHW kernelSize, Weights kernelWeights, Weights biasWeights) noexcept
Add a deconvolution layer to the network.
Definition: NvInfer.h:6395
IGatherLayer * addGatherV2(ITensor &data, ITensor &indices, GatherMode mode)
Add gather with specified mode, axis=0 and nbElementWiseDims=0.
Definition: NvInfer.h:6665
IShuffleLayer * addShuffle(ITensor &input) noexcept
Add a shuffle layer to the network.
Definition: NvInfer.h:6476
void setName(char const *name) noexcept
Sets the name of the network.
Definition: NvInfer.h:6901
ILRNLayer * addLRN(ITensor &input, int32_t window, float alpha, float beta, float k) noexcept
Add a LRN layer to the network.
Definition: NvInfer.h:6315
ITopKLayer * addTopK(ITensor &input, TopKOperation op, int32_t k, uint32_t reduceAxes) noexcept
Add a TopK layer to the network.
Definition: NvInfer.h:6633
IAssertionLayer * addAssertion(ITensor &condition, char const *message) noexcept
Add an assertion layer to the network.
Definition: NvInfer.h:7205
IScaleLayer * addScaleNd(ITensor &input, ScaleMode mode, Weights shift, Weights scale, Weights power, int32_t channelAxis) noexcept
Add a multi-dimension scale layer to the network.
Definition: NvInfer.h:7099
char const * getName() const noexcept
Returns the name associated with the network.
Definition: NvInfer.h:6915
TRT_DEPRECATED bool hasExplicitPrecision() const noexcept
True if network is an explicit precision network.
Definition: NvInfer.h:7132
IParametricReLULayer * addParametricReLU(ITensor &input, ITensor &slope) noexcept
Add a parametric ReLU layer to the network.
Definition: NvInfer.h:6999
ITensor * getOutput(int32_t index) const noexcept
Get the output tensor specified by the given index.
Definition: NvInfer.h:6560
ITensor * getInput(int32_t index) const noexcept
Get the input tensor specified by the given index.
Definition: NvInfer.h:6530
bool unmarkOutputForShapes(ITensor &tensor) noexcept
Undo markOutputForShapes.
Definition: NvInfer.h:6981
IFillLayer * addFill(Dims dimensions, FillOperation op) noexcept
Add a fill layer to the network.
Definition: NvInfer.h:7223
ILoop * addLoop() noexcept
Add a loop to the network.
Definition: NvInfer.h:7148
IDeconvolutionLayer * addDeconvolutionNd(ITensor &input, int32_t nbOutputMaps, Dims kernelSize, Weights kernelWeights, Weights biasWeights) noexcept
Add a multi-dimension deconvolution layer to the network.
Definition: NvInfer.h:7063
IActivationLayer * addActivation(ITensor &input, ActivationType type) noexcept
Add an activation layer to the network.
Definition: NvInfer.h:6277
virtual ~INetworkDefinition() noexcept=default
ILayer * getLayer(int32_t index) const noexcept
Get the layer specified by the given index.
Definition: NvInfer.h:6502
TRT_DEPRECATED IRNNv2Layer * addRNNv2(ITensor &input, int32_t layerCount, int32_t hiddenSize, int32_t maxSeqLen, RNNOperation op) noexcept
Add an layerCount deep RNN layer to the network with hiddenSize internal states that can take a batch...
Definition: NvInfer.h:6799
TRT_DEPRECATED IFullyConnectedLayer * addFullyConnected(ITensor &input, int32_t nbOutputs, Weights kernelWeights, Weights biasWeights) noexcept
Add a fully connected layer to the network.
Definition: NvInfer.h:6257
IIfConditional * addIfConditional() noexcept
Add an If-conditional layer to the network.
Definition: NvInfer.h:7367
IErrorRecorder * getErrorRecorder() const noexcept
get the ErrorRecorder assigned to this interface.
Definition: NvInfer.h:7294
int32_t getNbInputs() const noexcept
Get the number of inputs in the network.
Definition: NvInfer.h:6514
bool hasImplicitBatchDimension() const noexcept
Query whether the network was created with an implicit batch dimension.
Definition: NvInfer.h:6951
IReduceLayer * addReduce(ITensor &input, ReduceOperation operation, uint32_t reduceAxes, bool keepDimensions) noexcept
Add a reduce layer to the network.
Definition: NvInfer.h:6599
IUnaryLayer * addUnary(ITensor &input, UnaryOperation operation) noexcept
Add a unary layer to the network.
Definition: NvInfer.h:6445
void removeTensor(ITensor &tensor) noexcept
remove a tensor from the network definition.
Definition: NvInfer.h:6829
ISelectLayer * addSelect(ITensor &condition, ITensor &thenInput, ITensor &elseInput) noexcept
Add a select layer to the network.
Definition: NvInfer.h:7188
IScatterLayer * addScatter(ITensor &data, ITensor &indices, ITensor &updates, ScatterMode mode) noexcept
Add a Scatter layer to the network with specified mode and axis=0.
Definition: NvInfer.h:7333
int32_t getNbLayers() const noexcept
Get the number of layers in the network.
Definition: NvInfer.h:6488
apiv::VNetworkDefinition * mImpl
Definition: NvInfer.h:7387
IPoolingLayer * addPoolingNd(ITensor &input, PoolingType type, Dims windowSize) noexcept
Add a multi-dimension pooling layer to the network.
Definition: NvInfer.h:7041
bool markOutputForShapes(ITensor &tensor) noexcept
Enable tensor's value to be computed by IExecutionContext::getShapeBinding.
Definition: NvInfer.h:6969
TRT_DEPRECATED IPaddingLayer * addPadding(ITensor &input, DimsHW prePadding, DimsHW postPadding) noexcept
Add a padding layer to the network.
Definition: NvInfer.h:6462
IScaleLayer * addScale(ITensor &input, ScaleMode mode, Weights shift, Weights scale, Weights power) noexcept
Add a Scale layer to the network.
Definition: NvInfer.h:6342
void unmarkOutput(ITensor &tensor) noexcept
unmark a tensor as a network output.
Definition: NvInfer.h:6841
IIdentityLayer * addIdentity(ITensor &input) noexcept
Add an identity layer.
Definition: NvInfer.h:6814
TRT_DEPRECATED IConvolutionLayer * addConvolution(ITensor &input, int32_t nbOutputMaps, DimsHW kernelSize, Weights kernelWeights, Weights biasWeights) noexcept
Add a convolution layer to the network.
Definition: NvInfer.h:6234
IQuantizeLayer * addQuantize(ITensor &input, ITensor &scale) noexcept
Add a quantization layer to the network.
Definition: NvInfer.h:7352
IElementWiseLayer * addElementWise(ITensor &input1, ITensor &input2, ElementWiseOperation op) noexcept
Add an elementwise layer to the network.
Definition: NvInfer.h:6423
void setErrorRecorder(IErrorRecorder *recorder) noexcept
Set the ErrorRecorder for this interface.
Definition: NvInfer.h:7279
ISliceLayer * addSlice(ITensor &input, Dims start, Dims size, Dims stride) noexcept
Add a slice layer to the network.
Definition: NvInfer.h:6879
IConstantLayer * addConstant(Dims dimensions, Weights weights) noexcept
Add a constant layer to the network.
Definition: NvInfer.h:6730
IRaggedSoftMaxLayer * addRaggedSoftMax(ITensor &input, ITensor &bounds) noexcept
Add a RaggedSoftMax layer to the network.
Definition: NvInfer.h:6683
IShapeLayer * addShape(ITensor &input) noexcept
Add a shape layer to the network.
Definition: NvInfer.h:6933
IGatherLayer * addGather(ITensor &data, ITensor &indices, int32_t axis) noexcept
Add gather with mode GatherMode::kDEFAULT and specified axis and nbElementWiseDims=0.
Definition: NvInfer.h:6649
TRT_DEPRECATED IPoolingLayer * addPooling(ITensor &input, PoolingType type, DimsHW windowSize) noexcept
Add a pooling layer to the network.
Definition: NvInfer.h:6296
TRT_DEPRECATED void destroy() noexcept
Destroy this INetworkDefinition object.
Definition: NvInfer.h:6572
IResizeLayer * addResize(ITensor &input) noexcept
Add a resize layer to the network.
Definition: NvInfer.h:7115
IMatrixMultiplyLayer * addMatrixMultiply(ITensor &input0, MatrixOperation op0, ITensor &input1, MatrixOperation op1) noexcept
Add a MatrixMultiply layer to the network.
Definition: NvInfer.h:6704
ISoftMaxLayer * addSoftMax(ITensor &input) noexcept
Add a SoftMax layer to the network.
Definition: NvInfer.h:6355
IEinsumLayer * addEinsum(ITensor *const *inputs, int32_t nbInputs, char const *equation) noexcept
Add an Einsum layer to the network.
Definition: NvInfer.h:7381
void markOutput(ITensor &tensor) noexcept
Mark a tensor as a network output.
Definition: NvInfer.h:6211
int32_t getNbOutputs() const noexcept
Get the number of outputs in the network.
Definition: NvInfer.h:6544
TRT_DEPRECATED IPaddingLayer * addPaddingNd(ITensor &input, Dims prePadding, Dims postPadding) noexcept
Add a padding layer to the network. Only 2D padding is currently supported.
Definition: NvInfer.h:7240
bool setWeightsName(Weights weights, char const *name) noexcept
Associate a name with all current uses of the given weights.
Definition: NvInfer.h:7260
Forward declaration of IEngineInspector for use by other interfaces.
Definition: NvInferRuntime.h:43
Optimization profile for dynamic input dimensions and shape tensors.
Definition: NvInferRuntime.h:1120
Layer that represents a padding operation.
Definition: NvInfer.h:3801
Dims getPostPaddingNd() const noexcept
Get the padding that is applied at the end of the tensor.
Definition: NvInfer.h:3902
TRT_DEPRECATED DimsHW getPrePadding() const noexcept
Get the padding that is applied at the start of the tensor.
Definition: NvInfer.h:3824
TRT_DEPRECATED void setPostPadding(DimsHW padding) noexcept
Set the padding that is applied at the end of the tensor.
Definition: NvInfer.h:3838
virtual ~IPaddingLayer() noexcept=default
TRT_DEPRECATED void setPrePadding(DimsHW padding) noexcept
Set the padding that is applied at the start of the tensor.
Definition: NvInfer.h:3812
Dims getPrePaddingNd() const noexcept
Get the padding that is applied at the start of the tensor.
Definition: NvInfer.h:3876
TRT_DEPRECATED DimsHW getPostPadding() const noexcept
Get the padding that is applied at the end of the tensor.
Definition: NvInfer.h:3850
void setPostPaddingNd(Dims padding) noexcept
Set the padding that is applied at the end of the tensor.
Definition: NvInfer.h:3890
void setPrePaddingNd(Dims padding) noexcept
Set the padding that is applied at the start of the tensor.
Definition: NvInfer.h:3864
apiv::VPaddingLayer * mImpl
Definition: NvInfer.h:3908
Layer that represents a parametric ReLU operation.
Definition: NvInfer.h:4641
apiv::VParametricReLULayer * mImpl
Definition: NvInfer.h:4643
virtual ~IParametricReLULayer() noexcept=default
Single registration point for all plugins in an application. It is used to find plugin implementation...
Definition: NvInferRuntimeCommon.h:1210
Plugin class for user-implemented layers.
Definition: NvInferRuntimeCommon.h:373
Layer type for pluginV2.
Definition: NvInfer.h:3572
virtual ~IPluginV2Layer() noexcept=default
apiv::VPluginV2Layer * mImpl
Definition: NvInfer.h:3585
IPluginV2 & getPlugin() noexcept
Get the plugin for the layer.
Definition: NvInfer.h:3579
A Pooling layer in a network definition.
Definition: NvInfer.h:1666
TRT_DEPRECATED DimsHW getStride() const noexcept
Get the stride for pooling.
Definition: NvInfer.h:1739
PoolingType getPoolingType() const noexcept
Get the type of activation to be performed.
Definition: NvInfer.h:1685
void setWindowSizeNd(Dims windowSize) noexcept
Set the multi-dimension window size for pooling.
Definition: NvInfer.h:1918
PaddingMode getPaddingMode() const noexcept
Get the padding mode.
Definition: NvInfer.h:1905
Dims getPostPadding() const noexcept
Get the padding.
Definition: NvInfer.h:1881
TRT_DEPRECATED void setStride(DimsHW stride) noexcept
Set the stride for pooling.
Definition: NvInfer.h:1727
bool getAverageCountExcludesPadding() const noexcept
Get whether average pooling uses as a denominator the overlap area between the window and the unpadde...
Definition: NvInfer.h:1825
Dims getPrePadding() const noexcept
Get the pre-padding.
Definition: NvInfer.h:1853
void setPoolingType(PoolingType type) noexcept
Set the type of activation to be performed.
Definition: NvInfer.h:1675
void setPaddingMode(PaddingMode paddingMode) noexcept
Set the padding mode.
Definition: NvInfer.h:1894
TRT_DEPRECATED void setWindowSize(DimsHW windowSize) noexcept
Set the window size for pooling.
Definition: NvInfer.h:1699
Dims getWindowSizeNd() const noexcept
Get the multi-dimension window size for pooling.
Definition: NvInfer.h:1928
void setAverageCountExcludesPadding(bool exclusive) noexcept
Set whether average pooling uses as a denominator the overlap area between the window and the unpadde...
Definition: NvInfer.h:1814
void setPaddingNd(Dims padding) noexcept
Set the multi-dimension padding for pooling.
Definition: NvInfer.h:1972
void setPrePadding(Dims padding) noexcept
Set the multi-dimension pre-padding for pooling.
Definition: NvInfer.h:1843
float getBlendFactor() const noexcept
Get the blending factor for the max_average_blend mode: max_average_blendPool = (1-blendFactor)*maxPo...
Definition: NvInfer.h:1797
Dims getStrideNd() const noexcept
Get the multi-dimension stride for pooling.
Definition: NvInfer.h:1953
virtual ~IPoolingLayer() noexcept=default
Dims getPaddingNd() const noexcept
Get the multi-dimension padding for pooling.
Definition: NvInfer.h:1984
TRT_DEPRECATED DimsHW getWindowSize() const noexcept
Get the window size for pooling.
Definition: NvInfer.h:1711
TRT_DEPRECATED DimsHW getPadding() const noexcept
Get the padding for pooling.
Definition: NvInfer.h:1769
void setStrideNd(Dims stride) noexcept
Set the multi-dimension stride for pooling.
Definition: NvInfer.h:1943
void setPostPadding(Dims padding) noexcept
Set the multi-dimension post-padding for pooling.
Definition: NvInfer.h:1871
TRT_DEPRECATED void setPadding(DimsHW padding) noexcept
Set the padding for pooling.
Definition: NvInfer.h:1755
void setBlendFactor(float blendFactor) noexcept
Set the blending factor for the max_average_blend mode: max_average_blendPool = (1-blendFactor)*maxPo...
Definition: NvInfer.h:1784
A Quantize layer in a network definition.
Definition: NvInfer.h:5829
void setAxis(int32_t axis) noexcept
Set the quantization axis.
Definition: NvInfer.h:5850
int32_t getAxis() const noexcept
Get the quantization axis.
Definition: NvInfer.h:5839
virtual ~IQuantizeLayer() noexcept=default
An RNN layer in a network definition, version 2.
Definition: NvInfer.h:3282
void setDirection(RNNDirection op) noexcept
Set the direction of the RNN layer.
Definition: NvInfer.h:3382
void setBiasForGate(int32_t layerIndex, RNNGateType gate, bool isW, Weights bias) noexcept
Set the bias parameters for an individual gate in the RNN.
Definition: NvInfer.h:3486
void setCellState(ITensor &cell) noexcept
Set the initial cell state of the LSTM with the provided cell ITensor.
Definition: NvInfer.h:3542
int32_t getDataLength() const noexcept
Get the maximum data length of the RNN.
Definition: NvInfer.h:3296
void setInputMode(RNNInputMode op) noexcept
Set the input mode of the RNN layer.
Definition: NvInfer.h:3357
Weights getBiasForGate(int32_t layerIndex, RNNGateType gate, bool isW) const noexcept
Get the bias parameters for an individual gate in the RNN.
Definition: NvInfer.h:3496
void setWeightsForGate(int32_t layerIndex, RNNGateType gate, bool isW, Weights weights) noexcept
Set the weight parameters for an individual gate in the RNN.
Definition: NvInfer.h:3451
void setOperation(RNNOperation op) noexcept
Set the operation of the RNN layer.
Definition: NvInfer.h:3337
Weights getWeightsForGate(int32_t layerIndex, RNNGateType gate, bool isW) const noexcept
Get the weight parameters for an individual gate in the RNN.
Definition: NvInfer.h:3461
RNNDirection getDirection() const noexcept
Get the direction of the RNN layer.
Definition: NvInfer.h:3392
void setSequenceLengths(ITensor &seqLengths) noexcept
Specify individual sequence lengths in the batch with the ITensor pointed to by seqLengths.
Definition: NvInfer.h:3315
RNNInputMode getInputMode() const noexcept
Get the input mode of the RNN layer.
Definition: NvInfer.h:3367
ITensor * getHiddenState() const noexcept
Get the initial hidden state of the RNN.
Definition: NvInfer.h:3523
ITensor * getCellState() const noexcept
Get the initial cell state of the RNN.
Definition: NvInfer.h:3552
int32_t getMaxSeqLength() const noexcept
Get the maximum sequence length of the RNN.
Definition: NvInfer.h:3292
int32_t getLayerCount() const noexcept
Get the layer count of the RNN.
Definition: NvInfer.h:3284
apiv::VRNNv2Layer * mImpl
Definition: NvInfer.h:3558
ITensor * getSequenceLengths() const noexcept
Get the sequence lengths specified for the RNN.
Definition: NvInfer.h:3327
RNNOperation getOperation() const noexcept
Get the operation of the RNN layer.
Definition: NvInfer.h:3347
virtual ~IRNNv2Layer() noexcept=default
int32_t getHiddenSize() const noexcept
Get the hidden size of the RNN.
Definition: NvInfer.h:3288
void setHiddenState(ITensor &hidden) noexcept
Set the initial hidden state of the RNN with the provided hidden ITensor.
Definition: NvInfer.h:3513
A RaggedSoftmax layer in a network definition.
Definition: NvInfer.h:4531
apiv::VRaggedSoftMaxLayer * mImpl
Definition: NvInfer.h:4533
virtual ~IRaggedSoftMaxLayer() noexcept=default
Definition: NvInfer.h:5271
virtual ~IRecurrenceLayer() noexcept=default
Layer that represents a reduction across a non-bool tensor.
Definition: NvInfer.h:3723
void setKeepDimensions(bool keepDimensions) noexcept
Set the boolean that specifies whether or not to keep the reduced dimensions for the layer.
Definition: NvInfer.h:3770
void setOperation(ReduceOperation op) noexcept
Set the reduce operation for the layer.
Definition: NvInfer.h:3730
ReduceOperation getOperation() const noexcept
Get the reduce operation for the layer.
Definition: NvInfer.h:3740
virtual ~IReduceLayer() noexcept=default
uint32_t getReduceAxes() const noexcept
Get the axes over which to reduce for the layer.
Definition: NvInfer.h:3760
void setReduceAxes(uint32_t reduceAxes) noexcept
Set the axes over which to reduce.
Definition: NvInfer.h:3750
apiv::VReduceLayer * mImpl
Definition: NvInfer.h:3786
bool getKeepDimensions() const noexcept
Get the boolean that specifies whether or not to keep the reduced dimensions for the layer.
Definition: NvInfer.h:3780
A resize layer in a network definition.
Definition: NvInfer.h:4829
void setSelectorForSinglePixel(ResizeSelector selector) noexcept
Set coordinate selector function when resized to single pixel.
Definition: NvInfer.h:5018
void setOutputDimensions(Dims dimensions) noexcept
Set the output dimensions.
Definition: NvInfer.h:4849
void setNearestRounding(ResizeRoundMode value) noexcept
Set rounding mode for nearest neighbor resize.
Definition: NvInfer.h:5042
virtual ~IResizeLayer() noexcept=default
int32_t getScales(int32_t size, float *scales) const noexcept
Copies resize scales to scales[0, ..., nbScales-1], where nbScales is the number of scales that were ...
Definition: NvInfer.h:4908
ResizeMode getResizeMode() const noexcept
Get resize mode for an input tensor.
Definition: NvInfer.h:4930
void setScales(float const *scales, int32_t nbScales) noexcept
Set the resize scales.
Definition: NvInfer.h:4889
void setResizeMode(ResizeMode resizeMode) noexcept
Set resize mode for an input tensor.
Definition: NvInfer.h:4920
ResizeSelector getSelectorForSinglePixel() const noexcept
Get the coordinate selector function when resized to single pixel.
Definition: NvInfer.h:5028
TRT_DEPRECATED bool getAlignCorners() const noexcept
True if align corners has been set.
Definition: NvInfer.h:4958
void setCoordinateTransformation(ResizeCoordinateTransformation coordTransform) noexcept
Set coordinate transformation function.
Definition: NvInfer.h:4993
Dims getOutputDimensions() const noexcept
Get the output dimensions.
Definition: NvInfer.h:4859
ResizeRoundMode getNearestRounding() const noexcept
Get rounding mode for nearest neighbor resize.
Definition: NvInfer.h:5052
TRT_DEPRECATED void setAlignCorners(bool alignCorners) noexcept
Set whether to align corners while resizing.
Definition: NvInfer.h:4946
ResizeCoordinateTransformation getCoordinateTransformation() const noexcept
Get coordinate transformation function.
Definition: NvInfer.h:5003
A Scale layer in a network definition.
Definition: NvInfer.h:2148
Weights getScale() const noexcept
Get the scale value.
Definition: NvInfer.h:2205
Weights getPower() const noexcept
Get the power value.
Definition: NvInfer.h:2225
void setScale(Weights scale) noexcept
Set the scale value.
Definition: NvInfer.h:2195
void setPower(Weights power) noexcept
Set the power value.
Definition: NvInfer.h:2215
ScaleMode getMode() const noexcept
Get the scale mode.
Definition: NvInfer.h:2165
void setShift(Weights shift) noexcept
Set the shift value.
Definition: NvInfer.h:2175
void setChannelAxis(int32_t channelAxis) noexcept
Set the channel axis.
Definition: NvInfer.h:2261
Weights getShift() const noexcept
Get the shift value.
Definition: NvInfer.h:2185
virtual ~IScaleLayer() noexcept=default
void setMode(ScaleMode mode) noexcept
Set the scale mode.
Definition: NvInfer.h:2155
int32_t getChannelAxis() const noexcept
Get the channel axis.
Definition: NvInfer.h:2240
A scatter layer in a network definition. Supports several kinds of scattering.
Definition: NvInfer.h:6093
void setMode(ScatterMode mode) noexcept
Set the scatter mode.
Definition: NvInfer.h:6100
apiv::VScatterLayer * mImpl
Definition: NvInfer.h:6134
void setAxis(int32_t axis) noexcept
Set the axis used by ScatterMode::kELEMENTS.
Definition: NvInfer.h:6120
int32_t getAxis() const noexcept
Get the axis.
Definition: NvInfer.h:6128
ScatterMode getMode() const noexcept
Get the scatter mode.
Definition: NvInfer.h:6110
virtual ~IScatterLayer() noexcept=default
Definition: NvInfer.h:5519
virtual ~ISelectLayer() noexcept=default
Layer type for getting shape of a tensor.
Definition: NvInfer.h:4315
virtual ~IShapeLayer() noexcept=default
apiv::VShapeLayer * mImpl
Definition: NvInfer.h:4317
Layer type for shuffling data.
Definition: NvInfer.h:3936
apiv::VShuffleLayer * mImpl
Definition: NvInfer.h:4091
void setReshapeDimensions(Dims dimensions) noexcept
Set the reshaped dimensions.
Definition: NvInfer.h:3984
void setFirstTranspose(Permutation permutation) noexcept
Set the permutation applied by the first transpose operation.
Definition: NvInfer.h:3947
void setSecondTranspose(Permutation permutation) noexcept
Set the permutation applied by the second transpose operation.
Definition: NvInfer.h:4044
Dims getReshapeDimensions() const noexcept
Get the reshaped dimensions.
Definition: NvInfer.h:3997
Permutation getFirstTranspose() const noexcept
Get the permutation applied by the first transpose operation.
Definition: NvInfer.h:3959
virtual ~IShuffleLayer() noexcept=default
Permutation getSecondTranspose() const noexcept
Get the permutation applied by the second transpose operation.
Definition: NvInfer.h:4056
bool getZeroIsPlaceholder() const noexcept
Get meaning of 0 in reshape dimensions.
Definition: NvInfer.h:4085
void setZeroIsPlaceholder(bool zeroIsPlaceholder) noexcept
Set meaning of 0 in reshape dimensions.
Definition: NvInfer.h:4072
Slices an input tensor into an output tensor based on the offset and strides.
Definition: NvInfer.h:4165
void setMode(SliceMode mode) noexcept
Set the slice mode.
Definition: NvInfer.h:4259
apiv::VSliceLayer * mImpl
Definition: NvInfer.h:4298
virtual ~ISliceLayer() noexcept=default
void setStride(Dims stride) noexcept
Set the stride for computing the output slice data.
Definition: NvInfer.h:4234
void setStart(Dims start) noexcept
Set the start offset that the slice layer uses to create the output slice.
Definition: NvInfer.h:4176
Dims getStart() const noexcept
Get the start offset for the slice layer.
Definition: NvInfer.h:4191
void setSize(Dims size) noexcept
Set the dimensions of the output slice.
Definition: NvInfer.h:4205
Dims getSize() const noexcept
Get dimensions of the output slice.
Definition: NvInfer.h:4220
SliceMode getMode() const noexcept
Get the slice mode.
Definition: NvInfer.h:4269
Dims getStride() const noexcept
Get the stride for the output slice.
Definition: NvInfer.h:4249
A Softmax layer in a network definition.
Definition: NvInfer.h:2293
void setAxes(uint32_t axes) noexcept
Set the axis along which softmax is computed. Currently, only one axis can be set.
Definition: NvInfer.h:2325
uint32_t getAxes() const noexcept
Get the axis along which softmax occurs.
Definition: NvInfer.h:2335
virtual ~ISoftMaxLayer() noexcept=default
A tensor in a network definition.
Definition: NvInfer.h:167
bool setDynamicRange(float min, float max) noexcept
Set dynamic range for the tensor.
Definition: NvInfer.h:267
void setAllowedFormats(TensorFormats formats) noexcept
Set allowed formats for this tensor. By default all formats are allowed. Shape tensors (for which isS...
Definition: NvInfer.h:403
TensorLocation getLocation() const noexcept
Get the storage location of a tensor.
Definition: NvInfer.h:331
void resetDynamicRange() noexcept
Undo effect of setDynamicRange.
Definition: NvInfer.h:364
void setName(char const *name) noexcept
Set the tensor name.
Definition: NvInfer.h:181
bool isExecutionTensor() const noexcept
Whether the tensor is an execution tensor.
Definition: NvInfer.h:474
void setType(DataType type) noexcept
Set the data type of a tensor.
Definition: NvInfer.h:240
bool dynamicRangeIsSet() const noexcept
Query whether dynamic range is set.
Definition: NvInfer.h:356
void setLocation(TensorLocation location) noexcept
Set the storage location of a tensor.
Definition: NvInfer.h:346
char const * getName() const noexcept
Get the tensor name.
Definition: NvInfer.h:193
bool isShapeTensor() const noexcept
Whether the tensor is a shape tensor.
Definition: NvInfer.h:451
float getDynamicRangeMax() const noexcept
Get maximum of dynamic range.
Definition: NvInfer.h:384
bool isNetworkInput() const noexcept
Whether the tensor is a network input.
Definition: NvInfer.h:275
bool isNetworkOutput() const noexcept
Whether the tensor is a network output.
Definition: NvInfer.h:283
bool getBroadcastAcrossBatch() const noexcept
Check if tensor is broadcast across the batch.
Definition: NvInfer.h:321
void setBroadcastAcrossBatch(bool broadcastAcrossBatch) noexcept
Set whether to enable broadcast of tensor across the batch.
Definition: NvInfer.h:305
DataType getType() const noexcept
Get the data type of a tensor.
Definition: NvInfer.h:252
apiv::VTensor * mImpl
Definition: NvInfer.h:480
float getDynamicRangeMin() const noexcept
Get minimum of dynamic range.
Definition: NvInfer.h:374
virtual ~ITensor() noexcept=default
void setDimensions(Dims dimensions) noexcept
Set the dimensions of a tensor.
Definition: NvInfer.h:212
Dims getDimensions() const noexcept
Get the dimensions of a tensor.
Definition: NvInfer.h:225
TensorFormats getAllowedFormats() const noexcept
Get a bitmask of TensorFormat values that the tensor supports. For a shape tensor,...
Definition: NvInfer.h:416
Class to handle tactic timing info collected from builder.
Definition: NvInfer.h:7963
bool combine(ITimingCache const &inputCache, bool ignoreMismatch) noexcept
Combine input timing cache into local instance.
Definition: NvInfer.h:8000
virtual ~ITimingCache() noexcept=default
apiv::VTimingCache * mImpl
Definition: NvInfer.h:8016
bool reset() noexcept
Empty the timing cache.
Definition: NvInfer.h:8010
Layer that represents a TopK reduction.
Definition: NvInfer.h:4351
void setK(int32_t k) noexcept
Set the k value for the layer.
Definition: NvInfer.h:4380
void setReduceAxes(uint32_t reduceAxes) noexcept
Set which axes to reduce for the layer.
Definition: NvInfer.h:4400
TopKOperation getOperation() const noexcept
Get the operation for the layer.
Definition: NvInfer.h:4368
apiv::VTopKLayer * mImpl
Definition: NvInfer.h:4416
void setOperation(TopKOperation op) noexcept
Set the operation for the layer.
Definition: NvInfer.h:4358
int32_t getK() const noexcept
Get the k value for the layer.
Definition: NvInfer.h:4390
uint32_t getReduceAxes() const noexcept
Get the axes to reduce for the layer.
Definition: NvInfer.h:4410
virtual ~ITopKLayer() noexcept=default
Definition: NvInfer.h:5374
TripLimit getTripLimit() const noexcept
Definition: NvInfer.h:5376
virtual ~ITripLimitLayer() noexcept=default
Layer that represents an unary operation.
Definition: NvInfer.h:3648
void setOperation(UnaryOperation op) noexcept
Set the unary operation for the layer.
Definition: NvInfer.h:3657
apiv::VUnaryLayer * mImpl
Definition: NvInfer.h:3673
UnaryOperation getOperation() const noexcept
Get the unary operation for the layer.
Definition: NvInfer.h:3667
virtual ~IUnaryLayer() noexcept=default
An array of weights used as a layer parameter.
Definition: NvInferRuntime.h:126
The TensorRT API version 1 namespace.
uint32_t TacticSources
Represents a collection of one or more TacticSource values combine using bitwise-OR operations.
Definition: NvInferRuntime.h:1305
ResizeSelector
The coordinate selector when resize to single pixel output.
Definition: NvInfer.h:4734
@ kFORMULA
Use formula to map the original index.
@ kUPPER
Select the upper left pixel.
EngineCapability
List of supported engine capability flows.
Definition: NvInferRuntime.h:69
nvinfer1::IPluginRegistry * getBuilderPluginRegistry(nvinfer1::EngineCapability capability) noexcept
Return the plugin registry for the given capability or nullptr if no registry exists.
MemoryPoolType
The type for memory pools used by TensorRT.
Definition: NvInfer.h:8027
ScaleMode
Controls how shift, scale and power are applied in a Scale layer.
Definition: NvInfer.h:2104
@ kUNIFORM
Identical coefficients across all elements of the tensor.
@ kCHANNEL
Per-channel coefficients.
uint32_t QuantizationFlags
Represents one or more QuantizationFlag values using binary OR operations.
Definition: NvInfer.h:7844
constexpr int32_t EnumMax< RNNDirection >() noexcept
Definition: NvInfer.h:3204
constexpr int32_t EnumMax< BuilderFlag >() noexcept
Definition: NvInfer.h:7947
constexpr int32_t EnumMax< LayerType >() noexcept
Definition: NvInfer.h:103
constexpr int32_t EnumMax< RNNGateType >() noexcept
Definition: NvInfer.h:3265
constexpr int32_t EnumMax< CalibrationAlgoType >() noexcept
Definition: NvInfer.h:7409
UnaryOperation
Enumerates the unary operations that may be performed by a Unary layer.
Definition: NvInfer.h:3603
@ kCOSH
Hyperbolic cosine.
@ kACOSH
Inverse hyperbolic cosine.
@ kERF
Gauss error function.
@ kROUND
Round to nearest even for float datatype.
@ kATANH
Inverse hyperbolic tangent.
@ kASINH
Inverse hyperbolic sine.
@ kSIGN
Sign, If input > 0, output 1; if input < 0, output -1; if input == 0, output 0.
constexpr int32_t EnumMax< ReduceOperation >() noexcept
Definition: NvInfer.h:3710
constexpr int32_t EnumMax< TripLimit >() noexcept
Definition: NvInfer.h:5100
constexpr int32_t EnumMax< RNNInputMode >() noexcept
Definition: NvInfer.h:3236
RNNInputMode
Enumerates the RNN input modes that may occur with an RNN layer.
Definition: NvInfer.h:3225
@ kSKIP
No operation is performed on the first recurrent layer.
@ kLINEAR
Perform the normal matrix multiplication in the first recurrent layer.
ActivationType
Enumerates the types of activation to perform in an activation layer.
Definition: NvInfer.h:122
@ kSELU
Selu activation: x>0 ? beta * x : beta * (alpha*exp(x) - alpha)
@ kSCALED_TANH
Scaled tanh activation: alpha*tanh(beta*x)
@ kRELU
Rectified linear activation.
@ kELU
Elu activation: x>=0 ? x : alpha * (exp(x) - 1).
@ kLEAKY_RELU
LeakyRelu activation: x>=0 ? x : alpha * x.
@ kSOFTSIGN
Softsign activation: x / (1+|x|)
@ kHARD_SIGMOID
Hard sigmoid activation: max(0, min(1, alpha*x+beta))
@ kTHRESHOLDED_RELU
Thresholded ReLU activation: x>alpha ? x : 0.
@ kSIGMOID
Sigmoid activation.
@ kCLIP
Clip activation: max(alpha, min(beta, x))
@ kSOFTPLUS
Parametric softplus activation: alpha*log(exp(beta*x)+1)
FillOperation
Enumerates the tensor fill operations that may performed by a fill layer.
Definition: NvInfer.h:5579
@ kLINSPACE
Generate evenly spaced numbers over a specified interval.
@ kRANDOM_UNIFORM
Generate a tensor with random values drawn from a uniform distribution.
ResizeRoundMode
The rounding mode for nearest neighbor resize.
Definition: NvInfer.h:4764
@ kHALF_DOWN
Round half down.
RNNGateType
Identifies an individual gate within an RNN cell.
Definition: NvInfer.h:3249
@ kUPDATE
Update gate (z).
@ kHIDDEN
Hidden gate (h).
@ kFORGET
Forget gate (f).
@ kOUTPUT
Output gate (o).
PaddingMode
Enumerates the modes of padding to perform in convolution, deconvolution and pooling layer,...
Definition: NvInfer.h:951
@ kSAME_LOWER
Use SAME padding, with prePadding >= postPadding.
@ kEXPLICIT_ROUND_DOWN
Use explicit padding, rounding output size down.
@ kCAFFE_ROUND_DOWN
Use CAFFE padding, rounding output size down, uses prePadding value.
@ kEXPLICIT_ROUND_UP
Use explicit padding, rounding output size up.
@ kCAFFE_ROUND_UP
Use CAFFE padding, rounding output size up, uses prePadding value.
@ kSAME_UPPER
Use SAME padding, with prePadding <= postPadding.
TripLimit
Enum that describes kinds of trip limits.
Definition: NvInfer.h:5088
@ kWHILE
Tensor is a scalar of type kBOOL. Loop terminates when value is false.
@ kCOUNT
Tensor is scalar of type kINT32 that contains the trip count.
uint32_t NetworkDefinitionCreationFlags
Represents one or more NetworkDefinitionCreationFlag flags using binary OR operations....
Definition: NvInfer.h:8738
constexpr int32_t EnumMax< GatherMode >() noexcept
Definition: NvInfer.h:2920
DataType
The type of weights and tensors.
Definition: NvInferRuntimeCommon.h:114
uint32_t BuilderFlags
Represents one or more QuantizationFlag values using binary OR operations, e.g., 1U << BuilderFlag::k...
Definition: NvInfer.h:7878
DeviceType
The device that this layer/network will execute on.
Definition: NvInferRuntime.h:598
constexpr int32_t EnumMax< ScaleMode >() noexcept
Definition: NvInfer.h:2116
CalibrationAlgoType
Version of calibration algorithm to use.
Definition: NvInfer.h:7396
LayerType
The type values of layer classes.
Definition: NvInfer.h:53
@ kRAGGED_SOFTMAX
Ragged softmax layer.
@ kDECONVOLUTION
Deconvolution layer.
@ kASSERTION
Assertion layer.
@ kMATRIX_MULTIPLY
Matrix multiply layer.
@ kCONDITION
Condition layer.
@ kCONDITIONAL_INPUT
Conditional Input layer.
@ kIDENTITY
Identity layer.
@ kQUANTIZE
Quantize layer.
@ kCONVOLUTION
Convolution layer.
@ kPARAMETRIC_RELU
Parametric ReLU layer.
@ kCONCATENATION
Concatenation layer.
@ kFULLY_CONNECTED
Fully connected layer.
@ kRECURRENCE
Loop Recurrence layer.
@ kDEQUANTIZE
Dequantize layer.
@ kITERATOR
Loop Iterator layer.
@ kTRIP_LIMIT
Loop Trip limit layer.
@ kUNARY
UnaryOp operation Layer.
@ kACTIVATION
Activation layer.
@ kELEMENTWISE
Elementwise layer.
@ kPLUGIN_V2
PluginV2 layer.
@ kLOOP_OUTPUT
Loop output layer.
@ kCONDITIONAL_OUTPUT
Conditional Output layer.
@ kCONSTANT
Constant layer.
SliceMode
Controls how ISliceLayer handles out of bounds coordinates.
Definition: NvInfer.h:4101
@ kCLAMP
Out of bounds indices are clamped to bounds.
@ kWRAP
Coordinates wrap around periodically.
constexpr int32_t EnumMax< QuantizationFlag >() noexcept
Definition: NvInfer.h:7867
GatherMode
Control form of IGatherLayer.
Definition: NvInfer.h:2908
@ kDEFAULT
Similar to ONNX Gather.
@ kELEMENT
Similar to ONNX GatherElements.
@ kND
Similar to ONNX GatherND.
uint32_t TensorFormats
It is capable of representing one or more TensorFormat by binary OR operations, e....
Definition: NvInfer.h:114
ProfilingVerbosity
List of verbosity levels of layer information exposed in NVTX annotations and in IEngineInspector.
Definition: NvInferRuntime.h:1317
ResizeMode
Enumerates various modes of resize in the resize layer. Resize mode set using setResizeMode().
Definition: NvInfer.h:4653
@ kNEAREST
ND (0 < N <= 8) nearest neighbor resizing.
constexpr int32_t EnumMax< SliceMode >() noexcept
Definition: NvInfer.h:4117
NetworkDefinitionCreationFlag
List of immutable network properties expressed at network creation time. NetworkDefinitionCreationFla...
Definition: NvInfer.h:8749
ElementWiseOperation
Enumerates the binary operations that may be performed by an ElementWise layer.
Definition: NvInfer.h:2818
@ kSUB
Subtract the second element from the first.
@ kSUM
Sum of the two elements.
@ kPROD
Product of the two elements.
@ kFLOOR_DIV
Floor division of the first element by the second.
@ kEQUAL
Check if two elements are equal.
@ kAND
Logical AND of two elements.
@ kOR
Logical OR of two elements.
@ kMIN
Minimum of the two elements.
@ kPOW
The first element to the power of the second element.
@ kLESS
Check if element in first tensor is less than corresponding element in second tensor.
@ kGREATER
Check if element in first tensor is greater than corresponding element in second tensor.
@ kXOR
Logical XOR of two elements.
@ kDIV
Divide the first element by the second.
QuantizationFlag
List of valid flags for quantizing the network to int8.
Definition: NvInfer.h:7854
@ kCALIBRATE_BEFORE_FUSION
RNNDirection
Enumerates the RNN direction that may be performed by an RNN layer.
Definition: NvInfer.h:3193
@ kBIDIRECTION
Network iterates from first to last and vice versa and outputs concatenated.
@ kUNIDIRECTION
Network iterations from first input to last input.
BuilderFlag
List of valid modes that the builder can enable when creating an engine from a network definition.
Definition: NvInfer.h:7888
@ kDEBUG
Enable debugging of layers via synchronizing after every layer.
@ kGPU_FALLBACK
Enable layers marked to execute on GPU if layer cannot execute on DLA.
@ kSPARSE_WEIGHTS
Allow the builder to examine weights and use optimized functions when weights have suitable sparsity.
@ kFP16
Enable FP16 layer selection, with FP32 fallback.
@ kINT8
Enable Int8 layer selection, with FP32 fallback with FP16 fallback if kFP16 also specified.
@ kPREFER_PRECISION_CONSTRAINTS
@ kDISABLE_TIMING_CACHE
Disable reuse of timing information across identical layers.
@ kREFIT
Enable building a refittable engine.
@ kOBEY_PRECISION_CONSTRAINTS
Require that layers execute in specified precisions. Build fails otherwise.
@ kREJECT_EMPTY_ALGORITHMS
Fail if IAlgorithmSelector::selectAlgorithms returns an empty set of algorithms.
constexpr int32_t EnumMax< TopKOperation >() noexcept
Definition: NvInfer.h:4338
TensorFormat
Format of the input/output tensors.
Definition: NvInferRuntimeCommon.h:183
constexpr int32_t EnumMax< MemoryPoolType >() noexcept
Definition: NvInfer.h:8066
TopKOperation
Enumerates the operations that may be performed by a TopK layer.
Definition: NvInfer.h:4327
ReduceOperation
Enumerates the reduce operations that may be performed by a Reduce layer.
Definition: NvInfer.h:3696
constexpr int32_t EnumMax< LoopOutput >() noexcept
Definition: NvInfer.h:5081
RNNOperation
Enumerates the RNN operations that may be performed by an RNN layer.
Definition: NvInfer.h:3167
@ kGRU
Three-gate network consisting of Gated Recurrent Units.
@ kLSTM
Four-gate LSTM network w/o peephole connections.
constexpr int32_t EnumMax< NetworkDefinitionCreationFlag >() noexcept
Definition: NvInfer.h:8768
ScatterMode
Control form of IScatterLayer.
Definition: NvInfer.h:6020
MatrixOperation
Enumerates the operations that may be performed on a tensor by IMatrixMultiplyLayer before multiplica...
Definition: NvInfer.h:4427
@ kTRANSPOSE
Like kNONE, but transpose the matrix dimensions.
ResizeCoordinateTransformation
The resize coordinate transformation function.
Definition: NvInfer.h:4680
constexpr int32_t EnumMax< UnaryOperation >() noexcept
Definition: NvInfer.h:3635
LoopOutput
Enum that describes kinds of loop outputs.
Definition: NvInfer.h:5064
@ kLAST_VALUE
Output value is value of tensor for last iteration.
@ kCONCATENATE
Output value is concatenation of values of tensor for each iteration, in forward order.
@ kREVERSE
Output value is concatenation of values of tensor for each iteration, in reverse order.
constexpr int32_t EnumMax< MatrixOperation >() noexcept
Definition: NvInfer.h:4455
constexpr int32_t EnumMax< RNNOperation >() noexcept
Definition: NvInfer.h:3180
PoolingType
The type of pooling to perform in a pooling layer.
Definition: NvInfer.h:1634
constexpr int32_t EnumMax< FillOperation >() noexcept
Definition: NvInfer.h:5590
TensorLocation
The location for tensor data storage, device or host.
Definition: NvInferRuntime.h:216
OptProfileSelector
When setting or querying optimization profile parameters (such as shape tensor inputs or dynamic dime...
Definition: NvInferRuntime.h:1080
constexpr int32_t EnumMax< ScatterMode >() noexcept
Definition: NvInfer.h:6031
Definition: NvInfer.h:3913
Declaration of EnumMaxImpl struct to store maximum number of elements in an enumeration type.
Definition: NvInferRuntimeCommon.h:99