neat Namespace

Definition

namespace simaai::neat { ... }

Namespaces Index

namespace	validators
namespace	genai
namespace	graphs
namespace	internal
namespace	nodes
namespace	debug
namespace	error_codes
namespace	graph
namespace	pipeline_internal
namespace	runtime
namespace	stages
namespace	dlpack
	Minimal DLPack-shaped structs for zero-copy interop with NumPy/PyTorch. More...

Classes Index

struct	ValidationContext
	Context passed to contracts during validation. More...
class	Contract
	A single validation rule applied to an ordered node list. More...
class	ContractRegistry
	Holds a set of Contracts and runs them to produce a ValidationReport. More...
struct	ContractPortSpec
	Per-port contract: the shape/type/segment requirements for one input or output. More...
struct	ContractFieldSpec
	Per-field contract: where the field value comes from and the override policy. More...
struct	NodeContractDefinition
	Bundle of port and field specs that fully describes a Node's contract. More...
struct	ValidationIssue
	A single reported issue from a Contract. More...
class	ValidationReport
	Report produced by running a ContractRegistry. More...
struct	MlaLayer
	Hardware latency of one compiled MLA layer or segment. More...
struct	MlaLayerStats
	MLA layer latencies from BenchmarkOptions::include_mla_layer_stats. More...
struct	MlaStage
	Hardware latency of one compiled MLA stage, run uninterrupted. More...
struct	MlaStageStats
	MLA stage latencies from BenchmarkOptions::include_mla_stage_stats. More...
struct	BenchmarkReport
	Headline result returned by Model::benchmark(). More...
struct	BenchmarkOptions
	Options for Model::benchmark() inputs and measurement. More...
class	Model
	Loaded form of a compiled model archive; the simplified entry point to run inference on Modalix. More...
struct	ResizeSpec
	Resize / letterbox / crop parameters. More...
struct	ColorConvertSpec
	Color-format conversion parameters. More...
struct	LayoutConvertSpec
	Axis-permutation parameters for HWC↔CHW-style layout conversions. More...
struct	NormalizeSpec
	Mean/stddev normalization parameters. More...
struct	QuantizeSpec
	INT8 (or other low-precision) quantization parameters. More...
struct	TessellateSpec
	MLA tessellation (tile-shuffle) parameters. More...
struct	PreprocessExplicitKnobs
	Per-stage flags recording which preprocess fields the user set explicitly. More...
struct	Transform
	Discriminated union of all per-stage transform specs. More...
struct	PreprocessOptions
	User-facing preprocess intent — what the application asks for. More...
struct	PreprocessContract
	Per-input contract describing what the preprocess stage expects to receive. More...
struct	PreprocessMetaContract
	Metadata-side contract — names and fields the preprocess stage requires. More...
struct	ResolvedPreprocessPlan
	Final resolved plan — what the framework actually compiled and runs. More...
struct	EncodedCapsFixupOptions
	Configuration for EncodedCapsFixup. More...
class	EncodedCapsFixup
	Identity node whose runtime probe fills invalid encoded caps fields. More...
class	FileInput
	Wraps GStreamer's filesrc element — reads encoded media from a path on disk. More...
class	ImageDecode
	Auto-detects and decodes still-image formats (JPEG, PNG, BMP, …) to raw video. More...
class	JpegDecode
	Decodes JPEG-encoded buffers to raw frames. More...
struct	JpegParseOptions
	Construction options for JpegParse. More...
class	JpegParse
	JPEG stream parser Node backed by GStreamer's jpegparse. More...
struct	MultipartHeaderCaptureOptions
	Selects which multipart part headers are captured as per-frame attributes. More...
struct	MultipartJpegDemuxOptions
	Construction options for MultipartJpegDemux. More...
class	MultipartJpegDemux
	Demux multipart HTTP streams into per-part buffers. More...
struct	OutputOptions
	Buffering / sync options for the Output Node. More...
class	Output
	Pull-side terminal sink. More...
struct	QueueOptions
	Optional limits and overflow behavior for a Queue node. More...
class	Queue
	Wraps GStreamer's queue element — inserts a thread/buffer boundary. More...
class	VideoConvert
	Wraps GStreamer's videoconvert. More...
class	VideoScale
	Wraps GStreamer's videoscale. More...
class	VideoTrackSelect
	Creates a qtdemux element and selects a specific video pad (e.g., demux.video_0). More...
struct	CameraInputOptions
	Options for CameraInput, a live libcamera/MIPI source. More...
class	CameraInput
struct	HttpSourceOptions
	Construction options for an HttpSource Node. More...
class	HttpSource
	HTTP source Node backed by GStreamer's souphttpsrc. More...
struct	PreprocessMetaTemplate
	Preprocess metadata template attached to ingress buffers. More...
struct	InputOptions
	Caps, buffering, and pool options for the Input Node. More...
class	Input
	Push-mode source Node. More...
struct	MetadataSenderOptions
struct	MetadataSenderSendOptions
struct	MetadataSenderStats
	Point-in-time diagnostics for one metadata sender. More...
class	MetadataSender
class	RTSPInput
	RTSP source Node — pulls a stream from an RTSP URL. More...
class	StillImageInput
	Source Node that emits a single image as a continuous video stream. More...
struct	UdpOutputOptions
	Destination + sync options for the UdpOutput Node. More...
class	UdpOutput
	UDP sink Node — writes packets to a UDP destination. More...
struct	UsbCameraInputOptions
	Options for UsbCameraInput, with CameraInputOptions' field names and defaults. More...
struct	UsbCameraInfo
	A USB camera as UsbCameraInput opens it. More...
class	H264CapsFixup
	Helper Node that fills in missing H.264 RTP caps fields with fallbacks. More...
class	H264Depacketize
	Depayloads RTP-encapsulated H.264 into a raw H.264 byte stream. More...
class	H265Depacketize
	Depayloads RTP-encapsulated H.265 into parsed Annex-B access units. More...
class	RTPJpegDepacketize
	Depayloads RTP-encapsulated JPEG frames into image/jpeg buffers. More...
struct	CastOptions
	Construction options for a Cast Node. More...
class	Cast
	CVU kernel Node that casts a tensor between FP32 and BF16 (no scale/zero-point). More...
struct	CastTessOptions
	Construction options for a CastTess Node. More...
class	CastTess
	Fused CVU kernel Node: Cast (FP32→BF16) followed by Tess. More...
struct	DequantOptions
	Construction options for a Dequant Node. More...
class	Dequant
	CVU kernel Node that dequantizes INT8 tensors to FP32 using scale + zero-point. More...
struct	DetessOptions
	Construction options for a Detess Node. More...
class	Detess
	CVU kernel Node that detessellates MLA-layout tensors back to natural HWC/CHW order. More...
struct	DetessCastOptions
	Construction options for a DetessCast Node. More...
class	DetessCast
	Fused CVU kernel Node: Detess followed by Cast (BF16→FP32). More...
struct	DetessDequantOptions
	Construction options for a DetessDequant Node. More...
class	DetessDequant
	Fused CVU kernel Node: Detess followed by Dequant (INT8→FP32). More...
class	H264Decode
	Hardware-accelerated H.264 decoder Node. More...
class	H264EncodeSima
	Hardware-accelerated H.264 encoder Node. More...
class	H264Packetize
	RTP H.264 payloader Node — wraps encoded H.264 into RTP packets. More...
struct	H264ParseOptions
	Construction options for an H264Parse Node — config-interval and optional caps lock. More...
class	H264Parse
	H.264 stream parser Node — parses NAL units, finds keyframes, optionally locks caps. More...
struct	ImageColorOptions
	YUV color interpretation. These settings do not perform transfer-function conversion. More...
struct	ResizeOptions
	Resize changes geometry only. Output is a raw image in the input format. More...
struct	ColorConvertOptions
	Color conversion preserves the fixed geometry negotiated for this run. More...
struct	NormalizationOptions
	UInt8 RGB/BGR/GRAY8 to dense BF16; coefficients are in semantic RGB order. More...
struct	PCIeSinkOptions
	Construction options for a PCIeSink Node. More...
class	PCIeSink
	Terminal sink Node that streams samples to a PCIe-connected host. More...
struct	PCIeSrcOptions
	Construction options for a PCIeSrc Node. More...
class	PCIeSrc
	Source-role Node that receives samples from a PCIe-connected host. More...
struct	PreprocOptions
	Construction options for a Preproc Node. More...
class	Preproc
	Fused CVU preprocessing Node — resize + color-convert + normalize (+ optional tess). More...
struct	QuantOptions
	Construction options for a Quant Node. More...
class	Quant
	CVU kernel Node that quantizes FP32 tensors to INT8 using scale + zero-point. More...
struct	QuantTessOptions
	Construction options for a QuantTess Node. More...
class	QuantTess
	Fused CVU kernel Node: Quant (FP32→INT8) followed by Tess. More...
struct	SimaArgMaxOptions
	Construction options for a SimaArgMax Node. More...
class	SimaArgMax
	A65 postprocess Node that emits the top-class index from a classification tensor. More...
struct	SimaBEVPoolOptions
	Construction options for a SimaBEVPool Node. More...
class	SimaBEVPool
	Kernel Node that pools camera depth and context features into a BEV grid. More...
struct	SimaBEVPoolPlanOptions
	Construction options for a SimaBEVPoolPlan Node. More...
class	SimaBEVPoolPlan
	Kernel Node that builds SimaBEVPool's plan from one frame's camera calibration. More...
class	SimaBoxDecode
	Postprocess Node that converts detection-head tensors into task-specific results. More...
struct	SimaDecodeOptions
	Configuration for SimaDecode. More...
class	SimaDecode
	Hardware-accelerated native decoder node. More...
struct	SimaEncodeOptions
	Codec-specific fields are optional; defaults are selected after type. More...
class	SimaEncode
	Hardware encoder with native 8-bit NV12 input preparation. More...
struct	SimaKernelQuantization
	Quantization fused into a kernel's store. More...
struct	SimaPillarScatterOptions
	Construction options for a SimaPillarScatter Node. More...
class	SimaPillarScatter
	Kernel Node that scatters pillar features onto a dense BEV canvas. More...
struct	SimaRenderOptions
	Construction options for a SimaRender Node. More...
class	SimaRender
	Node that renders bounding-box overlays onto a video frame. More...
struct	SimaVoxelizeOptions
	Construction options for a SimaVoxelize Node. More...
class	SimaVoxelize
	Kernel Node that voxelizes, decorates and quantizes one LiDAR sweep. More...
struct	TessOptions
	Construction options for a Tess Node. More...
class	Tess
	CVU kernel Node that tessellates a natural-layout tensor into MLA tile geometry. More...
struct	VisualFrontendCommonOptions
	Shared options for EV74 visual-frontend tensor Nodes. More...
struct	FeatureHistogramOptions
	8-bit grayscale histogram Node options. More...
struct	GriderFastOptions
	Grid-distributed FAST feature detector options. More...
struct	TrackDescriptorOptions
	FAST + BRIEF-style descriptor frontend options. More...
struct	TrackKLTOptions
	Pyramidal Lucas-Kanade / KLT tracker options. More...
struct	MetoakDepthOptions
	Metoak S315 depth: separated I420, unsigned disparity and calibration tensors to RGB, metric depth and XYZ, executed on EV74. More...
class	FeatureHistogram
class	GriderFast
class	TrackDescriptor
class	TrackKLT
class	MetoakDepth
struct	BoxDecodeOptions
struct	Box
	One axis-aligned detection produced by BoxDecode. More...
struct	BoxDecodeResult
	Parsed BoxDecode output paired with its raw byte buffer. More...
struct	OrientedBoxesHeaderV1
struct	OrientedBoxRecordV1
struct	OrientedDecodeTensors
	Owned dense CPU results. More...
struct	PoseDecodeTensors
	Decoded pose tensors for one BoxDecode output tensor. More...
struct	SegmentationDecodeTensors
	Decoded instance-segmentation tensors for one BoxDecode output tensor. More...
struct	SegmentationPoseDecodeTensors
	Decoded tensors for a BoxDecode output carrying masks and keypoints together. More...
class	ExecutionPlan
	Execution plan of one Model build, laid out as the graph it runs. More...
struct	FeaturePointsHeaderV1
	Fixed prefix of the FEATURE_POINTS_V1 payload. More...
struct	FeaturePointTensors
struct	FormatSpec
	Thin wrapper around FormatTag with implicit string conversions. More...
class	RtspServerHandle
	Live handle for a Graph running in RTSP server mode. More...
class	Graph
	The assembly stage — turns a list of Nodes into a runnable, deterministic pipeline. More...
struct	NodeLatencySummary
struct	GraphElementMetrics
struct	GraphNodeMetrics
struct	GraphLinkOptions
	Options for Graph::connect(from, to, options). More...
struct	VerboseOptions
	Per-topic framework verbosity controls. More...
struct	RtspServerOptions
	Options for Graph::run_rtsp() — controls the RTSP server's mount point and ports. More...
struct	ValidateOptions
	Options for Graph::validate(). More...
struct	ProcessCvuOptions
	Simple process-CVU backend placement for model pre/post stages. More...
struct	ProcessMlaOptions
	process-MLA execution options. More...
struct	PreparedRunnerOptions
	Retained source-compatible options for the retired prepared-route experiment. More...
struct	AdvancedExecutionOptions
	Intent-named, jargon-free execution controls (preferred over the raw legacy fields). More...
struct	GraphOptions
	Per-Graph construction options. More...
struct	OutputTensorOptions
	Options for Graph::add_output_tensor() — the tensor-friendly output helper. More...
struct	PullError
	Structured error returned by Run::pull() when status is Error. More...
struct	Sample
	Typed payload returned by Run::pull() and consumed by Run::push(). More...
struct	BusMessage
	A single GStreamer bus message captured during pipeline build/run. More...
struct	BoundaryFlowStats
	Per-boundary flow statistics — buffer counts at identity probes between Nodes. More...
struct	NodeReport
	Per-Node entry in the report, listing what that Node produced in the GStreamer pipeline. More...
struct	BuildAdaptationAction
	One line item from the build-adaptation log. More...
struct	BuildAdaptationSummary
	Snapshot of build-time adaptation state — present for build(input) flows. More...
struct	GraphReport
	Structured pipeline diagnostics — the framework's primary triage record. More...
class	NeatError
	Framework exception type carrying a structured GraphReport. More...
struct	PowerRailConfig
	One board power rail. More...
struct	PowerFieldReading
	One scalar rail reading. More...
struct	PowerRailReading
	Snapshot for one rail. More...
struct	PowerSnapshot
	Point-in-time board power snapshot. More...
struct	PowerMonitorOptions
	Options for a PowerMonitor. More...
struct	PowerFieldSummary
	Summary statistics for one scalar rail field. More...
struct	PowerRailSummary
	Summary statistics for one rail. More...
struct	PowerSummary
	Aggregated power monitor result. More...
class	PowerMonitor
	Collects Sentinel power samples over a measured interval. More...
struct	RunAdvancedOptions
	Advanced runtime tuning knobs (most users never set these). More...
struct	RunAutoExportOptions
	Optional build-time Run JSON export. More...
struct	RunOptions
	Per-Run runtime options. More...
struct	InputDropInfo
	Diagnostic record for a dropped input frame. More...
struct	MeasureOptions
	Options for framework-owned runtime measurement. More...
struct	MeasureLatencyStats
	Percentile summary for a measured latency series. More...
struct	MeasurePluginLatency
	Aggregated per-plugin/kernel timing captured during a measurement window. More...
struct	MeasureEdgeLatency
	Aggregated handoff/queue/transport timing between plugins or runtime nodes. More...
struct	MeasurePathStat
struct	MeasurePathIdentity
struct	MeasurePathNodeArrival
struct	MeasurePathInterPluginGap
struct	MeasurePathOutputTail
struct	MeasurePathTiming
struct	MeasureCounters
	Common input/output/drop counters captured during a measurement window. More...
struct	MeasureInputStats
	Input-boundary telemetry captured during a measurement window. More...
struct	MeasureReport
	Framework-owned report returned by MeasureScope::stop(). More...
class	MeasureScope
	Observation scope for measuring an application-owned push/pull interval. More...
class	Run
	Live pipeline handle: push inputs in, pull outputs out. More...
struct	RunExportOptions
	Options for exporting a built Run as graph JSON. More...
struct	SuperPointOptions
	SuperPoint-specific BoxDecode options. Negative spatial knobs mean "use profile/MPK default". More...
struct	TelemetryStats
	One Sentinel measurement summarized over a telemetry window. More...
struct	Telemetry
	Board telemetry Sentinel recorded during a window. More...
class	TelemetryMonitor
	Records board telemetry from Sentinel between start() and stop(). More...
struct	PcmAudio
struct	ConversionCost
	Cost estimate for a conversion: bytes moved + a coarse compute-class bucket (low/med/high). More...
struct	ConversionTrace
	Single audit-log entry for a conversion that occurred. More...
struct	ConversionTraceCollector
	Collects ConversionTrace entries for post-mortem auditing of a pipeline run. More...
struct	Device
	Device descriptor: type + numeric ID (for multi-device boards). More...
struct	ImageSpec
	Image-tensor metadata: pixel format and (optional) color space. More...
struct	AudioSpec
	Audio-tensor metadata: sample rate, channel count, interleaving. More...
struct	TokensSpec
	Token-tensor metadata for NLP-style tensors. More...
struct	TextSpec
	UTF-8 text tensor metadata. More...
struct	EncodedSpec
	Encoded-stream tensor metadata: which codec the bytes represent. More...
struct	ByteStreamSpec
	Opaque byte-stream tensor metadata. More...
struct	QuantSpec
	Quantization metadata for INT8/INT16 tensors. More...
struct	TessSpec
	Tessellation metadata — tile geometry for the MLA's tile-block layout. More...
struct	DetectionSpec
	Detection-decoder metadata — tags tensors that carry packed detection output. More...
struct	FeatureSpec
	Feature-extractor output metadata (for example SuperPoint feature points). More...
struct	PreprocessRoi
	One runtime ROI window consumed by preprocessing. More...
struct	PreprocessAffine
	Per-ROI 2×3 affine matrix from model/preprocessed coordinates to source-frame coordinates. More...
struct	PreprocessRuntimeMeta
	Per-buffer preprocessing context — the inverse-transform breadcrumb trail. More...
struct	Semantic
	Discriminated union of "what this tensor represents". More...
struct	Mapping
	Scoped read/write window into a TensorBuffer. More...
struct	Segment
	One named memory segment within a multi-segment tensor buffer (e.g., separate Y / UV planes). More...
struct	TensorBuffer
	Storage handle for a tensor — opaque container for one of four backing memory kinds. More...
struct	Plane
	One plane of a composite (multi-plane) tensor. More...
struct	Nv12View
	Non-owning view into NV12 pixel data: Y plane + interleaved UV plane. More...
struct	Nv12Mapped
	Bundles an NV12 view with the Mapping that keeps its underlying buffer alive. More...
struct	I420View
	Non-owning view into I420 pixel data: separate Y, U, V planes. More...
struct	I420Mapped
	Bundles an I420 view with the Mapping that keeps its underlying buffer alive. More...
struct	TensorRouteMeta
	Routing metadata that travels with a tensor through multi-output pipelines. More...
struct	Tensor
	Universal tensor type — a labeled box of numbers that flows between Nodes. More...
struct	TensorConstraint
	Declarative tensor contract — describes the shape/dtype/device/format a tensor must satisfy. More...
struct	YoloXSegPoseOptions
	YOLOX segmentation/pose settings; other BoxDecode types reject a nonempty class list. More...

Typedefs Index

using	BusMessageFn = void(*)(const char *type, const char *src, const std::string &line, void *user_data)
	Callback invoked for every bus message drained. More...
using	BusErrorFn = void(*)(const std::string &line, void *user_data)
	Callback invoked when a bus error message is encountered. More...
using	TensorSpec = TensorConstraint
	Tensor specification used by the Model API. More...
using	BoxDecodeResultList = std::vector< BoxDecodeResult >
	List form used by public stage APIs; even single-image decode results travel as a list. More...
using	OrientedDecodeTensorList = std::vector< OrientedDecodeTensors >
using	PoseDecodeTensorList = std::vector< PoseDecodeTensors >
using	SegmentationDecodeTensorList = std::vector< SegmentationDecodeTensors >
using	SegmentationPoseDecodeTensorList = std::vector< SegmentationPoseDecodeTensors >
using	FeaturePointTensorList = std::vector< FeaturePointTensors >
using	RuntimeNodeId = std::size_t
using	GraphRunAutoExportOptions = RunAutoExportOptions
	Backward-compatible name for the previous graph-run export option spelling. More...
using	GraphRunExportOptions = RunExportOptions
	Backward-compatible name for callers that still use the old graph-run terminology. More...
using	TensorList = std::vector< Tensor >
	Convenience alias for an ordered list of Tensors (multi-input/multi-output models). More...
using	Memory = TensorMemory
	Short public alias for memory placement in Tensor/Sample constructors. More...
using	ImageType = ImageSpec::PixelFormat
	Short public alias for image pixel format in Tensor/Sample constructors. More...
using	Storage = TensorBuffer
	Alias for backward-compatibility. New code prefers TensorBuffer directly. More...
using	SampleAttributes = std::map< std::string, std::string >
	Per-frame user attributes carried alongside a Sample. More...
using	ByteFormat = ByteStreamSpec::ByteFormat
	Convenience top-level alias for ByteStreamSpec::ByteFormat. More...

Enumerations Index

enum class	CastDirection { ... }
	Direction of the BF16/FP32 conversion performed by a Cast Node. More...
enum class	ImageFormat { ... }
	Raw UInt8 image formats. Numerical tensor dtypes are intentionally separate. More...
enum class	ResizeFit { ... }
enum class	ImageColorMatrix { ... }
enum class	ImageYuvRange { ... }
enum class	ImageChromaSiting { ... }
enum class	ImageSurfaceLayout { ... }
enum class	ResizeInterpolation { ... }
	Sampling algorithm shared by Model preprocessing and standalone Resize. More...
enum class	PreprocColorConversionPlacement { ... }
	Whether YUV-to-RGB conversion precedes image sampling. No transfer-function conversion. More...
enum class	PreprocFloatingConversion { ... }
	Legacy preserves BF16 truncation. More...
enum class	SimaDecodeType { ... }
	Encoded media type decoded by SimaDecode. More...
enum class	SimaEncodeType { ... }
	Encoded output type produced by SimaEncode. More...
enum class	FeatureValueDType : std::uint8_t { ... }
enum class	FeaturePointsLayout : std::uint8_t { ... }
enum class	GraphLinkPolicy { ... }
	Runtime policy for a connection between two public Graph fragments. More...
enum class	PayloadType { ... }
	Semantic family of a payload flowing through a Graph. More...
enum class	MetricsTraceSource { ... }
	Metrics trace backend selection for framework-owned measurements. More...
enum class	SuperPointProfile : std::int32_t { ... }
	Numerical contract used by BoxDecodeType::SuperPoint. More...
enum class	SuperPointOutputFormat : std::int32_t { ... }
	Byte representation emitted by the SuperPoint BoxDecode backend. More...
enum class	TensorMemory { ... }
	Placement for Tensor creators that materialize owned storage. More...
enum class	MapMode { ... }
	Access mode for TensorBuffer::map(). More...
enum class	PayloadType { ... }
	Semantic family of a payload flowing through a Graph. More...
enum class	MemoryContract { ... }
	How a Node (or Graph) wants buffer memory to be sourced/handed back. More...
enum class	CapsMemory { ... }
	Memory class to advertise in GStreamer caps. More...
enum class	ContractFieldSource { ... }
	Where the value of a contract field originates. More...
enum class	ContractOverridePolicy { ... }
	Whether the Builder may override a contract field at build time. More...
enum class	ValidationSeverity { ... }
	Severity level for validation issues. More...
enum class	AutoFlag { ... }
	Tri-state knob: let the planner decide, force on, or force off. More...
enum class	InputKind { ... }
	What kind of input the user is feeding the preprocess pipeline. More...
enum class	ResizeMode { ... }
	How the resize stage maps non-matching input dimensions to the model's input shape. More...
enum class	PreprocessColorFormat { ... }
	Color format hint for the color-convert stage. More...
enum class	NormalizePreset { ... }
	Common normalize presets — convenient shorthand for famous mean/stddev pairs. More...
enum class	PreprocessGraphFamily { ... }
	Which preprocess graph family the planner selected. More...
enum class	TransformType { ... }
	Which transformation a Transform represents — used in user-supplied transform lists. More...
enum class	CombinePolicy { ... }
	How a public Graph output combines multiple named producers. More...
enum class	InputMemoryPolicy { ... }
	Where ingress buffers should be allocated when the application pushes samples. More...
enum class	SimaBEVPoolStorage { ... }
	Byte order of a SimaBEVPool depth, context or BEV tensor. More...
enum class	SimaKernelTarget { ... }
	Hardware that executes a kernels-library Node. More...
enum class	SimaQuantRounding { ... }
	Integer rounding applied after the binary32 product of a fused quantization. More...
enum class	SimaBEVPoolPlanFormat { ... }
	Layout of a SimaBEVPoolPlan plan, which SimaBEVPool must be told. More...
enum class	BoxDecodeType : std::int32_t { ... }
	Decode families accepted by the BoxDecode backend. More...
enum class	BoxDecodeTypeOption : std::int32_t { ... }
	Tensor packing/layout option within a decode family. More...
enum class	BoxDecodeSourceStorage : std::int32_t { ... }
	Source byte layout of the head tensors feeding a BoxDecode stage. More...
enum class	FormatTag { ... }
	Identifies a media or tensor payload format. More...
enum class	VerbosityLevel { ... }
	Coarse-grained framework verbosity selector. More...
enum class	RunMode { ... }
	Internal timing mode a Run operates in. More...
enum class	SampleKind { ... }
	What kind of payload a Sample carries. More...
enum class	PullStatus { ... }
	Result status of Run::pull(). More...
enum class	PowerMonitorProfile { ... }
	Named board-level power monitor profiles. More...
enum class	RunPreset { ... }
	Convenience preset bundles for RunOptions. More...
enum class	OutputMemory { ... }
	How output Tensors relate to the underlying GStreamer buffers. More...
enum class	ConversionKind { ... }
	What kind of transformation a tensor conversion represents. More...
enum class	ConversionPolicy { ... }
	How strict the framework is about implicit tensor conversions. More...
enum class	DeviceType { ... }
	Where a tensor's data lives — which processor can read it directly. More...
enum class	StorageKind { ... }
	How a TensorBuffer's memory was acquired and how to access it. More...
enum class	PlaneRole { ... }
	Role of a plane within a composite (multi-plane) tensor. More...
enum class	TensorDType { ... }
	Element data type of a tensor. More...
enum class	TensorAxisSemantic : uint8_t { ... }
	Per-axis semantic tag — the long-term layout vocabulary. More...
enum class	TensorLayout { ... }
	Transition-only coarse layout token (HWC / CHW / HW). More...

Operators Index

std::ostream &	operator<< (std::ostream &os, const FormatSpec &spec)
	Stream-insert the canonical string form of spec. More...
std::ostream &	operator<< (std::ostream &out, const Telemetry &telemetry)
	Write telemetry as the YAML text Telemetry::save() writes to a .yaml file. More...

Functions Index

const char *	to_string (ValidationSeverity s)
	Stable string label for a ValidationSeverity (for reports/logs). More...
std::string	gst_message_to_string (GstMessage *msg)
	Format a single GStreamer bus message into a stable, one-line string. More...
void	drain_bus (GstElement *pipeline, BusMessageFn on_message, void *user_data)
	Drain pending messages from the pipeline's bus, invoking on_message for each. More...
void	throw_if_bus_error (GstElement *pipeline, BusMessageFn on_message, void *user_data, BusErrorFn on_error, void *error_user_data)
	Drain the bus and throw a NeatError if any error message is found. More...
bool	element_exists (const char *factory)
	Returns true if a GStreamer element factory with factory name is registered. More...
bool	element_property_exists (const char *factory, const char *property_name)
	Returns true if factory exists and exposes a property named property_name. More...
std::string	factory_plugin_path (const char *factory)
	Returns the filesystem path of the plugin .so providing factory, or empty if unknown. More...
void	require_element (const char *factory, const char *context)
	Throw a NeatError if factory is not registered. More...
void	require_tensordecoder (const char *context)
	Throw a NeatError if SiMa's tensordecoder element is not available. More...
void	gst_init_once ()
	Initialize GStreamer exactly once across the process. More...
void	prewarm_runtime ()
	Initialize NEAT's runtime substrate early. More...
std::future< void >	prewarm_runtime_async ()
	Start NEAT runtime prewarm on a background thread. More...
std::string	resolve_input_media_type (const InputOptions &opt)
	Resolve the internal caps media type requested by public InputOptions. More...
std::vector< std::string >	usb_camera_ids ()
	Stable IDs of the connected USB (UVC) cameras, for UsbCameraInputOptions::camera_name. More...
UsbCameraInfo	usb_camera_info (const UsbCameraInputOptions &opt)
	Validate opt, find its camera, check the mode and read one frame. More...
constexpr const char *	box_decode_source_storage_token (BoxDecodeSourceStorage storage)
	Stable lower-case token for an explicit source-storage selection (diagnostics/docs). More...
constexpr const char *	box_decode_type_token (BoxDecodeType type)
	Stable lower-case token for a decode family (used in caps, manifests, logs). More...
constexpr const char *	box_decode_type_option_token (BoxDecodeTypeOption option)
	Stable lower-case token for a decode-layout option. More...
constexpr bool	box_decode_type_is_yolo_family (BoxDecodeType type)
	True iff type is one of the YOLO-family detection or segmentation variants. More...
constexpr bool	box_decode_type_is_oriented (BoxDecodeType type)
	True iff type produces oriented floating-point boxes. More...
constexpr bool	box_decode_type_is_ssd_family (BoxDecodeType type)
	True iff type selects the SSD decode family. More...
constexpr bool	box_decode_type_is_segmentation (BoxDecodeType type)
	True iff type is a segmentation variant (carries a mask head). More...
constexpr bool	box_decode_type_is_pose (BoxDecodeType type)
	True iff type is a pose-estimation variant (carries keypoint heads). More...
constexpr const char *	box_decode_type_contract_summary (BoxDecodeType type)
	Human-readable summary of type's tensor contract, used in API/docs/error surfaces. More...
OrientedDecodeTensors	decode_obb_tensor (const Tensor &tensor)
	Strict parsing of a tensor explicitly tagged BBOX_OBB_V1; malformed payloads throw. More...
OrientedDecodeTensorList	decode_obb (const TensorList &tensors)
	Positional one-to-one list parsing; no axis-aligned BBOX fallback. More...
bool	detection_format_is_bbox (const std::string &format)
	True iff format names the plain BBOX detection wire format. More...
bool	detection_format_is_pose (const std::string &format)
	True iff format names a BoxDecode pose wire format. More...
bool	detection_format_is_segmentation (const std::string &format)
	True iff format names a BoxDecode segmentation wire format. More...
bool	detection_format_is_segmentation_pose (const std::string &format)
	True iff format is any BoxDecode format whose leading payload is BBOX records. More...
bool	detection_format_is_bbox_family (const std::string &format)
SegmentationPoseDecodeTensors	decode_segmentation_pose_tensor (const simaai::neat::Tensor &tensor, int img_w=0, int img_h=0, int top_k=0, bool strict=false)
	Decode a combined segmentation + pose BoxDecode payload. More...
SegmentationPoseDecodeTensorList	decode_segmentation_pose (const simaai::neat::TensorList &tensors, int img_w=0, int img_h=0, int top_k=0, bool strict=false)
std::vector< Box >	parse_bbox_bytes (const std::vector< uint8_t > &bytes, int img_w, int img_h, int expected_topk, bool strict)
	Parse a packed BBOX byte payload into typed Box records. More...
BoxDecodeResult	decode_bbox_tensor (const simaai::neat::Tensor &tensor, int img_w, int img_h, int expected_topk, bool strict)
	Decode a BBOX-format Tensor into a BoxDecodeResult. More...
simaai::neat::Tensor	boxes_to_tensor (const std::vector< Box > &boxes)
	Build a decoded-boxes Tensor from parsed Box records. More...
simaai::neat::TensorList	decode_bbox (const simaai::neat::TensorList &bbox_tensors, int img_w=0, int img_h=0, int top_k=0, bool strict=false)
	Decode a list of BBOX-format tensors into a list of decoded-boxes tensors. More...
PoseDecodeTensors	decode_pose_tensor (const simaai::neat::Tensor &tensor, int img_w=0, int img_h=0, int top_k=0, bool strict=false)
	Decode one BoxDecode pose payload into boxes and keypoints tensors. More...
PoseDecodeTensorList	decode_pose (const simaai::neat::TensorList &pose_tensors, int img_w=0, int img_h=0, int top_k=0, bool strict=false)
	Decode a list of BoxDecode pose tensors, positional 1:1. More...
SegmentationDecodeTensors	decode_segmentation_tensor (const simaai::neat::Tensor &tensor, int img_w=0, int img_h=0, int top_k=0, bool strict=false)
	Decode one BoxDecode segmentation payload into boxes and masks tensors. More...
SegmentationDecodeTensorList	decode_segmentation (const simaai::neat::TensorList &segmentation_tensors, int img_w=0, int img_h=0, int top_k=0, bool strict=false)
	Decode a list of BoxDecode segmentation tensors, positional 1:1. More...
void	tag_detection_format (simaai::neat::Tensor &tensor, std::string format)
	Tag a tensor as carrying a detection-decoder wire format. More...
std::string	read_detection_format (const simaai::neat::Tensor &tensor)
	Read the detection-format token from a tensor, if present. More...
void	tag_detection_format_in_sample (simaai::neat::Sample &sample)
	Walk a Sample tree and tag every recognized BBOX-payload Tensor with semantic.detection.format = "BBOX". More...
simaai::neat::EncodedSpec::Codec	caps_to_codec (const std::string &caps_string)
	Map a GStreamer-style caps string to an EncodedSpec::Codec value. More...
Sample	make_encoded_sample (std::vector< uint8_t > bytes, std::string caps_string, int64_t pts_ns=-1, int64_t dts_ns=-1, int64_t duration_ns=-1)
	Build an encoded Sample from raw bytes and caps metadata. More...
FeaturePointTensors	decode_superpoint_tensor (const Tensor &tensor)
	Parse a V1 tensor, or the explicit legacy V0 representation when its feature tag says so. More...
FeaturePointTensorList	decode_superpoint (const TensorList &tensors)
	Positional one-to-one list form of decode_superpoint_tensor. More...
void	tag_feature_format (Tensor &tensor, std::string format)
std::string	read_feature_format (const Tensor &tensor)
void	tag_feature_format_in_sample (Sample &sample)
const char *	format_tag_name (FormatTag tag)
	Stable, canonical string token for tag (empty string for Auto). More...
std::string	format_tag_to_string (FormatTag tag)
	std::string form of the canonical format token. More...
std::string	upper_copy_ascii (std::string value)
	ASCII upper-case copy of value (does not touch non-ASCII bytes). More...
bool	tensor_format_is_bf16_alias (std::string value)
	True iff value names BF16 (any of BF16, BFLOAT16, EVXX_BF16, EVXX_BFLOAT16). More...
bool	tensor_format_is_fp32_alias (std::string value)
	True iff value names FP32 (FP32, FLOAT32, EVXX_FLOAT32). More...
bool	tensor_format_is_int8_alias (std::string value)
	True iff value names INT8 (INT8, EVXX_INT8). More...
bool	tensor_format_is_int16_alias (std::string value)
	True iff value names INT16 (INT16, EVXX_INT16). More...
bool	tensor_format_is_int32_alias (std::string value)
	True iff value names INT32 (INT32, EVXX_INT32). More...
std::string	normalize_tensor_caps_format (std::string value)
	Normalize tensor format aliases to the canonical EVXX_* token. More...
std::string	normalize_caps_format_for_media (std::string media_type, std::string format)
	Normalize a caps-format string conditionally on its media type. More...
FormatTag	format_tag_from_string (const std::string &value)
	Parse a string token to a FormatTag; unknown values map to Auto. More...
bool	is_raw_video_format (FormatTag tag)
	True iff tag names a raw (uncompressed) video format. More...
bool	is_tensor_payload_format (FormatTag tag)
	True iff tag names a tensor payload (MLA, BBOX, ARGMAX, dtype kinds, EVXX aliases). More...
PayloadType	sample_payload_type (const Sample &sample)
	Return a Sample's standardized payload family, falling back from legacy media_type. More...
std::string	sample_media_type (const Sample &sample)
	Return the internal media type corresponding to a Sample's payload family. More...
Sample	make_tensor_sample (const std::string &port_name, simaai::neat::Tensor tensor)
	Construct a TensorSet-kind Sample wrapping a single Tensor for the named port. More...
Sample	make_bundle_sample (std::initializer_list< Sample > fields)
	Construct a Bundle-kind Sample whose payload is the given list of inner Samples. More...
bool	sample_is_multi_output (const Sample &sample)
	Returns true if the sample carries multiple outputs (Bundle, or TensorSet with size > 1). More...
bool	sample_has_tensor_list (const Sample &sample)
	Returns true if the sample is a TensorSet with at least one tensor. More...
TensorList &	sample_tensor_list (Sample &sample, const char *where=nullptr)
	Mutable reference to the sample's tensor list. Throws NeatError if kind != TensorSet. More...
const TensorList &	sample_tensor_list (const Sample &sample, const char *where=nullptr)
	Const reference to the sample's tensor list. Throws NeatError if kind != TensorSet. More...
Tensor &	require_single_tensor (Sample &sample, const char *where=nullptr)
	Returns the sample's single Tensor; throws if the sample carries 0 or >1 tensors. More...
const Tensor &	require_single_tensor (const Sample &sample, const char *where=nullptr)
	Const variant of require_single_tensor. More...
TensorList	tensors_from_sample (const Sample &sample, bool require_nonempty=true)
	Extract all tensors from any sample shape (Tensor, TensorSet, Bundle); flattens recursively. More...
Sample	sample_from_tensors (const TensorList &tensors)
	Construct a TensorSet sample from a list of tensors. More...
PayloadType	payload_type_from_media_type (std::string_view media_type)
	Convert an internal/legacy caps media string to the public payload family. More...
PayloadType	input_type_from_media_type (std::string_view media_type)
	Legacy helper name kept as a source-compatible bridge. More...
std::string	media_type_from_payload_type (PayloadType payload_type)
	Convert a payload family to the canonical internal media type when one exists. More...
std::string	run_to_json (const Run &run, const RunExportOptions &opt={}, std::string *err=nullptr)
	Serialize a built Run to sima.neat.graph_run JSON. Returns empty string on error. More...
std::string	run_to_json (const Run &run, const MeasureReport &report, const RunExportOptions &opt={}, std::string *err=nullptr)
	Serialize a built Run plus a measurement-window report to graph JSON. More...
bool	save_run_json (const Run &run, const std::string &path, const RunExportOptions &opt={}, std::string *err=nullptr)
	Atomically write run_to_json(run, opt) to path using path + ".tmp" then rename. More...
bool	save_run_json (const Run &run, const MeasureReport &report, const std::string &path, const RunExportOptions &opt={}, std::string *err=nullptr)
	Atomically write run_to_json(run, report, opt) to path. More...
std::string	graph_run_to_json (const Run &run, const GraphRunExportOptions &opt={}, std::string *err=nullptr)
	Compatibility wrapper for existing graph-run JSON call sites. More...
std::string	graph_run_to_json (const Run &run, const MeasureReport &report, const GraphRunExportOptions &opt={}, std::string *err=nullptr)
	Compatibility wrapper for measured graph-run JSON call sites. More...
bool	save_graph_run_json (const Run &run, const std::string &path, const GraphRunExportOptions &opt={}, std::string *err=nullptr)
	Compatibility wrapper for existing graph-run JSON call sites. More...
bool	save_graph_run_json (const Run &run, const MeasureReport &report, const std::string &path, const GraphRunExportOptions &opt={}, std::string *err=nullptr)
	Compatibility wrapper for measured graph-run JSON call sites. More...
constexpr const char *	superpoint_profile_token (SuperPointProfile profile)
constexpr const char *	superpoint_output_format_token (SuperPointOutputFormat format)
Tensor	from_gst_sample (GstSample *sample)
PcmAudio	tensor_to_pcm_audio (const Tensor &tensor)
ConversionCost	estimate_conversion_cost (ConversionKind kind, std::uint64_t bytes_copied)
	Estimate the cost (bytes + compute class) of a conversion of the given kind moving bytes_copied bytes. More...
bool	conversion_allowed (ConversionPolicy policy, ConversionKind kind)
	Returns true if the policy allows a conversion of this kind (without performing it). More...
std::shared_ptr< TensorBuffer >	make_cpu_owned_storage (std::size_t size_bytes)
	Allocate a CPU-owned heap buffer of size_bytes, initialized to zero. More...
std::shared_ptr< TensorBuffer >	make_cpu_external_storage (void *data, std::size_t size_bytes, std::shared_ptr< void > holder={}, bool read_only=true)
	Wrap a foreign CPU pointer as a TensorBuffer without copying. More...
bool	is_tessellated_int8_format (const std::string &fmt)
	True iff fmt names an INT8 tessellated tensor format string. More...
bool	is_tessellated_bf16_format (const std::string &fmt)
	True iff fmt names a BF16 tessellated tensor format string. More...
std::vector< std::string >	normalize_multipart_header_capture (const MultipartHeaderCaptureOptions &opt)
	Normalize and validate a header-capture allowlist. More...
std::vector< PowerRailConfig >	default_modalix_som_power_rails ()
	Default Modalix SOM PMIC rail table from the board measurement scripts. More...
std::vector< PowerRailConfig >	default_modalix_dvt_power_rails ()
	Default Modalix DVT PMIC rail table. More...
PowerMonitorProfile	detect_default_power_monitor_profile ()
	Detect the most likely built-in board power profile for the current host. More...
std::string	power_monitor_profile_name (PowerMonitorProfile profile)
	Human-readable name for a power monitor profile. More...
std::vector< PowerRailConfig >	power_rails_for_profile (PowerMonitorProfile profile)
	Resolve a built-in board profile to its rail table. More...
PowerMonitorOptions	board_power_monitor_options (int sample_interval_ms=100, PowerMonitorProfile profile=PowerMonitorProfile::Auto)
	Options that enable board power monitoring. More...
PowerMonitorOptions	modalix_som_power_monitor_options (int sample_interval_ms=100)
	Convenience options enabling Modalix SOM power monitoring. More...
PowerMonitorOptions	modalix_dvt_power_monitor_options (int sample_interval_ms=100)
	Convenience options enabling Modalix DVT power monitoring. More...
PowerSnapshot	read_power_snapshot (const PowerMonitorOptions &options)
	Read Sentinel's latest power sample. More...
std::string	format_power_summary (const PowerSummary &summary)
	Format a compact human-readable power summary. More...
std::string	power_summary_to_json (const PowerSummary &summary, int indent=2)
	Format power summary as a JSON object. More...

Variables Index

constexpr std::size_t	kMultipartHeaderCaptureMaxHeaders = 64U
	Maximum number of distinct header names one capture allowlist may select. More...
constexpr std::size_t	kMultipartHeaderCaptureMaxNameBytes = 128U
	Maximum length, in bytes, of a single selected header name. More...
constexpr std::size_t	kMultipartHeaderCaptureMaxLineBytes = 8U * 1024U
	Maximum length, in bytes, of one header line inside a multipart part header block. More...
constexpr std::size_t	kMultipartHeaderCaptureMaxBlockBytes = 64U * 1024U
	Maximum total size, in bytes, of one multipart part header block. More...
constexpr char	kDetectionFormatBbox[] = "BBOX"
	Canonical detection-format token for BoxDecode bounding-box payloads. More...
constexpr char	kDetectionFormatBboxPose[] = "BBOX_POSE"
	Canonical detection-format token for BoxDecode pose payloads. More...
constexpr char	kDetectionFormatBboxSegmentation[] = "BBOX_SEGMENTATION"
	Canonical detection-format token for BoxDecode instance-segmentation payloads. More...
constexpr char	kDetectionFormatBboxSegmentationPose[] = "BBOX_SEGMENTATION_POSE"
	Canonical detection-format token for BoxDecode payloads carrying masks AND keypoints. More...
constexpr char	kDetectionFormatBboxObbV1[] = "BBOX_OBB_V1"
	Versioned, little-endian oriented-box carrier; not an axis-aligned BBOX payload. More...
constexpr int64_t	kDecodedPoseKeypoints = 17
	Number of keypoints emitted by the BoxDecode pose wire format. More...
constexpr int64_t	kDecodedPoseColumns = 3
	Number of values per keypoint: x, y, visibility. More...
constexpr int64_t	kDecodedMaskWidth = 160
	Width of each BoxDecode instance mask. More...
constexpr int64_t	kDecodedMaskHeight = 160
	Height of each BoxDecode instance mask. More...
constexpr int64_t	kDecodedBoxColumns = 6
	Number of columns in a decoded-boxes tensor: x1, y1, x2, y2, score, class_id. More...
constexpr char	kFeatureFormatPointsV1[] = "FEATURE_POINTS_V1"
constexpr char	kFeatureFormatLegacyA65V0[] = "FEATURE_POINTS_LEGACY_A65_V0"
constexpr std::uint32_t	kFeaturePointsMagicV1 = 0x31565046U
static constexpr RuntimeNodeId	kInvalidRuntimeNodeId = static_cast<RuntimeNodeId>(-1)

Typedefs

BoxDecodeResultList

using simaai::neat::BoxDecodeResultList = std::vector<BoxDecodeResult>

List form used by public stage APIs; even single-image decode results travel as a list.

Definition at line 61 of file DetectionTypes.h.

BusErrorFn

using simaai::neat::BusErrorFn = void (*)(const std::string& line, void* user_data)

Callback invoked when a bus error message is encountered.

Definition at line 36 of file GstBusWatch.h.

BusMessageFn

using simaai::neat::BusMessageFn = void (*)(const char* type, const char* src, const std::string& line, void* user_data)

Callback invoked for every bus message drained.

Parameters

type	GStreamer message type as a stable string (e.g., "warning", "info").
src	Name of the element that posted the message, or empty.
line	Pre-formatted single-line summary of the message.
user_data	Opaque pointer the caller registered.

Definition at line 32 of file GstBusWatch.h.

FeaturePointTensorList

using simaai::neat::FeaturePointTensorList = std::vector<FeaturePointTensors>

Definition at line 64 of file FeatureTypes.h.

GraphRunAutoExportOptions

using simaai::neat::GraphRunAutoExportOptions = RunAutoExportOptions

Backward-compatible name for the previous graph-run export option spelling.

Definition at line 144 of file Run.h.

GraphRunExportOptions

using simaai::neat::GraphRunExportOptions = RunExportOptions

Backward-compatible name for callers that still use the old graph-run terminology.

Definition at line 45 of file RunExport.h.

ImageType

using simaai::neat::ImageType = ImageSpec::PixelFormat

Short public alias for image pixel format in Tensor/Sample constructors.

Definition at line 158 of file TensorCore.h.

Memory

using simaai::neat::Memory = TensorMemory

Short public alias for memory placement in Tensor/Sample constructors.

Definition at line 108 of file TensorCore.h.

OrientedDecodeTensorList

using simaai::neat::OrientedDecodeTensorList = std::vector<OrientedDecodeTensors>

Definition at line 119 of file DetectionTypes.h.

PoseDecodeTensorList

using simaai::neat::PoseDecodeTensorList = std::vector<PoseDecodeTensors>

Definition at line 180 of file DetectionTypes.h.

RuntimeNodeId

using simaai::neat::RuntimeNodeId = std::size_t

Definition at line 23 of file GraphMetrics.h.

SegmentationDecodeTensorList

using simaai::neat::SegmentationDecodeTensorList = std::vector<SegmentationDecodeTensors>

Definition at line 181 of file DetectionTypes.h.

SegmentationPoseDecodeTensorList

using simaai::neat::SegmentationPoseDecodeTensorList = std::vector<SegmentationPoseDecodeTensors>

Definition at line 182 of file DetectionTypes.h.

Storage

using simaai::neat::Storage = TensorBuffer

Alias for backward-compatibility. New code prefers TensorBuffer directly.

Definition at line 550 of file TensorCore.h.

TensorList

using simaai::neat::TensorList = std::vector<Tensor>

Convenience alias for an ordered list of Tensors (multi-input/multi-output models).

Definition at line 23 of file Tensor.h.

TensorSpec

using simaai::neat::TensorSpec = TensorConstraint

Tensor specification used by the Model API.

Definition at line 55 of file Model.h.

Enumerations

CastDirection

enum class simaai::neat::CastDirection

strong

Direction of the BF16/FP32 conversion performed by a Cast Node.

Enumeration values

Bf16ToFp32	Read BF16, emit FP32 (typical post-MLA path) (= 0)
Fp32ToBf16	Read FP32, emit BF16 (typical pre-MLA path) (= 1)

Definition at line 30 of file Cast.h.

FeaturePointsLayout

enum class simaai::neat::FeaturePointsLayout : std::uint8_t

strong

Enumeration values

StructureOfArrays

(= 1)

Definition at line 28 of file FeatureTypes.h.

FeatureValueDType

enum class simaai::neat::FeatureValueDType : std::uint8_t

strong

Enumeration values

Int8	(= 1)
BFloat16	(= 2)
Float32	(= 3)

Definition at line 22 of file FeatureTypes.h.

GraphLinkPolicy

enum class simaai::neat::GraphLinkPolicy

strong

Runtime policy for a connection between two public Graph fragments.

Enumeration values

Default	(= 0)
RealtimeLatestByStream

Default preserves lossless/backpressure semantics for ordinary one-to-one edges. If multiple producers connect to the same live public input, the framework promotes those edges to RealtimeLatestByStream automatically so users do not need app-local fan-in mutex code. RealtimeLatestByStream keeps producers non-blocking, retains only the latest frame per Sample::stream_id (or per source edge when the stream id is empty), and schedules ready streams fairly into the downstream graph. Ordinary Graph::build() automatically chooses fused lowering for eligible live fan-in.

Definition at line 56 of file GraphOptions.h.

ImageChromaSiting

enum class simaai::neat::ImageChromaSiting

strong

Enumeration values

Centered
Cosited

Definition at line 17 of file ImageTransform.h.

ImageColorMatrix

enum class simaai::neat::ImageColorMatrix

strong

Enumeration values

Bt601
Bt709

Definition at line 15 of file ImageTransform.h.

ImageFormat

enum class simaai::neat::ImageFormat

strong

Raw UInt8 image formats. Numerical tensor dtypes are intentionally separate.

Enumeration values

RGB
BGR
GRAY8
NV12
I420
NV16

Definition at line 13 of file ImageTransform.h.

ImageSurfaceLayout

enum class simaai::neat::ImageSurfaceLayout

strong

Enumeration values

Native	Tight packed rows or 16-byte YUV pitches; preserve padding holes
EncoderNv12	Encoder NV12 pitch/storage rows; initialize declared padding on EV every frame

Definition at line 18 of file ImageTransform.h.

ImageYuvRange

enum class simaai::neat::ImageYuvRange

strong

Enumeration values

Full
Limited

Definition at line 16 of file ImageTransform.h.

MapMode

enum class simaai::neat::MapMode

strong

Access mode for TensorBuffer::map().

Enumeration values

Read	Read-only map; framework may skip cache invalidation if known clean (= 0)
Write	Write-only map; framework may skip read-back; flushes cache on unmap
ReadWrite	Both directions; framework treats as full ownership transfer for the mapping's lifetime

Affects whether the framework treats the mapping as read or write.

Definition at line 127 of file TensorCore.h.

MetricsTraceSource

enum class simaai::neat::MetricsTraceSource

strong

Metrics trace backend selection for framework-owned measurements.

Enumeration values

Auto
Off
Lttng

Auto prefers the low-overhead LTTng collector. If LTTng is unavailable the report marks the requested metric class as unavailable; it does not silently fall back to a second metrics backend.

Definition at line 274 of file Run.h.

PayloadType

enum class simaai::neat::PayloadType

strong

Semantic family of a payload flowing through a Graph.

Enumeration values

Auto	Infer from tensor/sample metadata when possible (= 0)
Image	Decoded image pixels (video/x-raw internally)
Tensor	Tensor payload (application/vnd.simaai.tensor internally)
Encoded	Encoded video/byte-stream payloads such as H.264, H.265, or JPEG

PayloadType is intentionally neutral: the same enum describes an input option, a pushed/pulled Sample, and a boundary contract. GStreamer media strings are an internal transport detail derived from this enum at the adapter boundary.

Definition at line 20 of file PayloadType.h.

PayloadType

enum class simaai::neat::PayloadType

strong

Semantic family of a payload flowing through a Graph.

PayloadType is intentionally neutral: the same enum describes an input option, a pushed/pulled Sample, and a boundary contract. GStreamer media strings are an internal transport detail derived from this enum at the adapter boundary.

Definition at line 28 of file PayloadType.h.

PreprocColorConversionPlacement

enum class simaai::neat::PreprocColorConversionPlacement

strong

Whether YUV-to-RGB conversion precedes image sampling. No transfer-function conversion.

Enumeration values

AfterResize	(= 0)
BeforeResize	(= 1)

Definition at line 8 of file PreprocMath.h.

PreprocFloatingConversion

enum class simaai::neat::PreprocFloatingConversion

strong

Legacy preserves BF16 truncation.

Enumeration values

Legacy	(= 0)
NearestEven	(= 1)

NearestEven explicitly selects BF16 nearest-even. FP16 remains binary16 nearest-even in either mode.

Definition at line 11 of file PreprocMath.h.

ResizeFit

enum class simaai::neat::ResizeFit

strong

Enumeration values

Stretch
Contain

Definition at line 14 of file ImageTransform.h.

ResizeInterpolation

enum class simaai::neat::ResizeInterpolation

strong

Sampling algorithm shared by Model preprocessing and standalone Resize.

Enumeration values

Nearest
Bilinear
Bicubic
BoxAverage
BilinearAntialiased

Existing enumerator values are preserved.

Definition at line 6 of file PreprocMath.h.

SimaDecodeType

enum class simaai::neat::SimaDecodeType

strong

Encoded media type decoded by SimaDecode.

Enumeration values

H264	H.264 elementary stream (= 0)
AVC	Alias for H.264 (= H264)
JPEG	Single JPEG frame (= 1)
MJPEG	JPEG-frame video stream after framing/depacketization (= 2)
H265	H.265/HEVC Main 8-bit 4:2:0 elementary stream (= 3)
HEVC	Alias for H.265 (= H265)

Definition at line 19 of file SimaDecode.h.

SimaEncodeType

enum class simaai::neat::SimaEncodeType

strong

Encoded output type produced by SimaEncode.

Enumeration values

H264	(= 0)
AVC	(= H264)
H265	(= 1)
HEVC	(= H265)
MJPEG	(= 2)

Definition at line 19 of file SimaEncode.h.

SuperPointOutputFormat

enum class simaai::neat::SuperPointOutputFormat : std::int32_t

strong

Byte representation emitted by the SuperPoint BoxDecode backend.

Enumeration values

FeaturePointsV1	(= 1)
LegacyA65InterleavedV0	(= 2)

Definition at line 29 of file SuperPointTypes.h.

SuperPointProfile

enum class simaai::neat::SuperPointProfile : std::int32_t

strong

Numerical contract used by BoxDecodeType::SuperPoint.

Enumeration values

Auto	(= 0)
LightGlueV1	(= 1)
MagicLeapDemoV1	(= 2)
PaperBicubicV1	(= 3)
A65V1	(= 4)

Auto is a construction-time sentinel. It selects authoritative MPK metadata when present and otherwise resolves to A65V1; it is never inferred from tensor geometry or values and must not reach the runtime backend.

Definition at line 20 of file SuperPointTypes.h.

TensorMemory

enum class simaai::neat::TensorMemory

strong

Placement for Tensor creators that materialize owned storage.

Enumeration values

EV74	(= 0)
CPU
MLA
A65
Auto

The default for new owning creators is EV74: DDR allocated from the SiMa allocator in the EV74-addressable window (not EV74 VCCM). Use CPU for ordinary malloc-backed host tensors, or MLA for DMS0-backed tensors when the runtime allocator is available.

Definition at line 99 of file TensorCore.h.

Operators

operator<<()

std::ostream & simaai::neat::operator<< (std::ostream & os, const FormatSpec & spec)

inline

Stream-insert the canonical string form of spec.

Definition at line 347 of file FormatSpec.h.

operator<<()

std::ostream & simaai::neat::operator<< (std::ostream & out, const Telemetry & telemetry)

Write telemetry as the YAML text Telemetry::save() writes to a .yaml file.

Definition at line 95 of file TelemetryMonitor.h.

Functions

box_decode_source_storage_token()

const char * simaai::neat::box_decode_source_storage_token (BoxDecodeSourceStorage storage)

constexpr

Stable lower-case token for an explicit source-storage selection (diagnostics/docs).

Definition at line 132 of file BoxDecodeType.h.

box_decode_type_contract_summary()

const char * simaai::neat::box_decode_type_contract_summary (BoxDecodeType type)

constexpr

Human-readable summary of type's tensor contract, used in API/docs/error surfaces.

Definition at line 354 of file BoxDecodeType.h.

box_decode_type_is_oriented()

bool simaai::neat::box_decode_type_is_oriented (BoxDecodeType type)

constexpr

True iff type produces oriented floating-point boxes.

Definition at line 272 of file BoxDecodeType.h.

box_decode_type_is_pose()

bool simaai::neat::box_decode_type_is_pose (BoxDecodeType type)

constexpr

True iff type is a pose-estimation variant (carries keypoint heads).

Definition at line 317 of file BoxDecodeType.h.

box_decode_type_is_segmentation()

bool simaai::neat::box_decode_type_is_segmentation (BoxDecodeType type)

constexpr

True iff type is a segmentation variant (carries a mask head).

Definition at line 282 of file BoxDecodeType.h.

box_decode_type_is_ssd_family()

bool simaai::neat::box_decode_type_is_ssd_family (BoxDecodeType type)

constexpr

True iff type selects the SSD decode family.

Definition at line 277 of file BoxDecodeType.h.

box_decode_type_is_yolo_family()

bool simaai::neat::box_decode_type_is_yolo_family (BoxDecodeType type)

constexpr

True iff type is one of the YOLO-family detection or segmentation variants.

Definition at line 235 of file BoxDecodeType.h.

box_decode_type_option_token()

const char * simaai::neat::box_decode_type_option_token (BoxDecodeTypeOption option)

constexpr

Stable lower-case token for a decode-layout option.

Definition at line 207 of file BoxDecodeType.h.

box_decode_type_token()

const char * simaai::neat::box_decode_type_token (BoxDecodeType type)

constexpr

Stable lower-case token for a decode family (used in caps, manifests, logs).

Definition at line 144 of file BoxDecodeType.h.

boxes_to_tensor()

simaai::neat::Tensor simaai::neat::boxes_to_tensor (const std::vector< Box > & boxes)

Build a decoded-boxes Tensor from parsed Box records.

Produces a dense CPU float32 tensor of shape [boxes.size(), 6]. Columns are x1, y1, x2, y2, score, class_id (class_id stored as float; exact for the integer range).

Definition at line 250 of file DetectionTypes.h.

caps_to_codec()

simaai::neat::EncodedSpec::Codec simaai::neat::caps_to_codec (const std::string & caps_string)

Map a GStreamer-style caps string to an EncodedSpec::Codec value.

Parameters

caps_string

GStreamer caps string (e.g., "video/x-h264, ...").

Returns

Matching codec enum, or the codec-unspecified sentinel if no match.

Definition at line 31 of file EncodedSampleUtil.h.

conversion_allowed()

bool simaai::neat::conversion_allowed (ConversionPolicy policy, ConversionKind kind)

Returns true if the policy allows a conversion of this kind (without performing it).

Definition at line 95 of file TensorConversion.h.

decode_bbox()

simaai::neat::TensorList simaai::neat::decode_bbox (const simaai::neat::TensorList & bbox_tensors, int img_w=0, int img_h=0, int top_k=0, bool strict=false)

Decode a list of BBOX-format tensors into a list of decoded-boxes tensors.

Positional, 1:1 — output [i] is the decode of input [i]. Each output is a dense CPU float32 tensor of shape [num_detections, 6] (columns x1, y1, x2, y2, score, class_id).

Every input tensor must be BBOX-format (semantic.detection.format == "BBOX"); a tensor that is not raises (no silent skip).

Parameters

bbox_tensors	The BBOX-format tensors (e.g. a model's run output).
img_w	Optional source width to clamp coordinates to (0 = no clamp).
img_h	Optional source height to clamp coordinates to (0 = no clamp).
top_k	Optional cap on detections per tensor (0 = no cap).
strict	When true, malformed buffers throw instead of best-effort parsing.

Returns

A TensorList of the same length as bbox_tensors.

Definition at line 268 of file DetectionTypes.h.

decode_bbox_tensor()

BoxDecodeResult simaai::neat::decode_bbox_tensor (const simaai::neat::Tensor & tensor, int img_w, int img_h, int expected_topk, bool strict)

Decode a BBOX-format Tensor into a BoxDecodeResult.

Convenience wrapper that maps the tensor, parses bytes via parse_bbox_bytes, and retains the raw byte view alongside the typed boxes.

Parameters

tensor	Tensor whose payload is a BBOX byte stream.
img_w	Original image width used to clamp/scale coordinates.
img_h	Original image height used to clamp/scale coordinates.
expected_topk	Number of detections expected (advisory; see strict).
strict	If true, mismatches against expected_topk throw.

Returns

Parsed detections plus retained raw bytes.

Definition at line 238 of file DetectionTypes.h.

decode_obb()

OrientedDecodeTensorList simaai::neat::decode_obb (const TensorList & tensors)

Positional one-to-one list parsing; no axis-aligned BBOX fallback.

Definition at line 123 of file DetectionTypes.h.

decode_obb_tensor()

OrientedDecodeTensors simaai::neat::decode_obb_tensor (const Tensor & tensor)

Strict parsing of a tensor explicitly tagged BBOX_OBB_V1; malformed payloads throw.

Definition at line 121 of file DetectionTypes.h.

decode_pose()

PoseDecodeTensorList simaai::neat::decode_pose (const simaai::neat::TensorList & pose_tensors, int img_w=0, int img_h=0, int top_k=0, bool strict=false)

Decode a list of BoxDecode pose tensors, positional 1:1.

Definition at line 284 of file DetectionTypes.h.

decode_pose_tensor()

PoseDecodeTensors simaai::neat::decode_pose_tensor (const simaai::neat::Tensor & tensor, int img_w=0, int img_h=0, int top_k=0, bool strict=false)

Decode one BoxDecode pose payload into boxes and keypoints tensors.

The source wire layout is the existing BoxDecode pose payload: uint32 count + top_k × BBOX + top_k × PoseOut. The returned boxes tensor has shape [N, 6]; the returned keypoints tensor has shape [N, 17, 3].

Definition at line 278 of file DetectionTypes.h.

decode_segmentation()

SegmentationDecodeTensorList simaai::neat::decode_segmentation (const simaai::neat::TensorList & segmentation_tensors, int img_w=0, int img_h=0, int top_k=0, bool strict=false)

Decode a list of BoxDecode segmentation tensors, positional 1:1.

Definition at line 303 of file DetectionTypes.h.

decode_segmentation_pose()

SegmentationPoseDecodeTensorList simaai::neat::decode_segmentation_pose (const simaai::neat::TensorList & tensors, int img_w=0, int img_h=0, int top_k=0, bool strict=false)

Definition at line 208 of file DetectionTypes.h.

decode_segmentation_pose_tensor()

SegmentationPoseDecodeTensors simaai::neat::decode_segmentation_pose_tensor (const simaai::neat::Tensor & tensor, int img_w=0, int img_h=0, int top_k=0, bool strict=false)

Decode a combined segmentation + pose BoxDecode payload.

Same contract as decode_segmentation_tensor and decode_pose_tensor, for a payload that carries both extension regions.

Definition at line 204 of file DetectionTypes.h.

decode_segmentation_tensor()

SegmentationDecodeTensors simaai::neat::decode_segmentation_tensor (const simaai::neat::Tensor & tensor, int img_w=0, int img_h=0, int top_k=0, bool strict=false)

Decode one BoxDecode segmentation payload into boxes and masks tensors.

The source wire layout is the existing BoxDecode segmentation payload: uint32 count + top_k × BBOX + top_k × 160 × 160 uint8 masks. The returned boxes tensor has shape [N, 6]; the returned masks tensor has shape [N, 160, 160].

Definition at line 295 of file DetectionTypes.h.

decode_superpoint()

FeaturePointTensorList simaai::neat::decode_superpoint (const TensorList & tensors)

Positional one-to-one list form of decode_superpoint_tensor.

Definition at line 70 of file FeatureTypes.h.

decode_superpoint_tensor()

FeaturePointTensors simaai::neat::decode_superpoint_tensor (const Tensor & tensor)

Parse a V1 tensor, or the explicit legacy V0 representation when its feature tag says so.

Definition at line 67 of file FeatureTypes.h.

detection_format_is_bbox()

bool simaai::neat::detection_format_is_bbox (const std::string & format)

True iff format names the plain BBOX detection wire format.

Definition at line 185 of file DetectionTypes.h.

detection_format_is_bbox_family()

bool simaai::neat::detection_format_is_bbox_family (const std::string & format)

Definition at line 196 of file DetectionTypes.h.

detection_format_is_pose()

bool simaai::neat::detection_format_is_pose (const std::string & format)

True iff format names a BoxDecode pose wire format.

Definition at line 188 of file DetectionTypes.h.

detection_format_is_segmentation()

bool simaai::neat::detection_format_is_segmentation (const std::string & format)

True iff format names a BoxDecode segmentation wire format.

Definition at line 191 of file DetectionTypes.h.

detection_format_is_segmentation_pose()

bool simaai::neat::detection_format_is_segmentation_pose (const std::string & format)

True iff format is any BoxDecode format whose leading payload is BBOX records.

Definition at line 194 of file DetectionTypes.h.

drain_bus()

void simaai::neat::drain_bus (GstElement * pipeline, BusMessageFn on_message, void * user_data)

Drain pending messages from the pipeline's bus, invoking on_message for each.

Non-blocking: pulls all messages currently posted, then returns. Safe to call from the framework's polling loop while the pipeline is running.

Parameters

pipeline	Pipeline whose bus to drain.
on_message	Callback invoked for each message (may be null).
user_data	Opaque pointer forwarded to on_message.

Definition at line 55 of file GstBusWatch.h.

element_exists()

bool simaai::neat::element_exists (const char * factory)

Returns true if a GStreamer element factory with factory name is registered.

Definition at line 20 of file GstHelpers.h.

element_property_exists()

bool simaai::neat::element_property_exists (const char * factory, const char * property_name)

Returns true if factory exists and exposes a property named property_name.

Definition at line 23 of file GstHelpers.h.

estimate_conversion_cost()

ConversionCost simaai::neat::estimate_conversion_cost (ConversionKind kind, std::uint64_t bytes_copied)

Estimate the cost (bytes + compute class) of a conversion of the given kind moving bytes_copied bytes.

Definition at line 93 of file TensorConversion.h.

factory_plugin_path()

std::string simaai::neat::factory_plugin_path (const char * factory)

Returns the filesystem path of the plugin .so providing factory, or empty if unknown.

Definition at line 26 of file GstHelpers.h.

format_tag_from_string()

FormatTag simaai::neat::format_tag_from_string (const std::string & value)

inline

Parse a string token to a FormatTag; unknown values map to Auto.

Definition at line 209 of file FormatSpec.h.

format_tag_name()

const char * simaai::neat::format_tag_name (FormatTag tag)

inline

Stable, canonical string token for tag (empty string for Auto).

Definition at line 67 of file FormatSpec.h.

format_tag_to_string()

std::string simaai::neat::format_tag_to_string (FormatTag tag)

inline

std::string form of the canonical format token.

Definition at line 120 of file FormatSpec.h.

from_gst_sample()

Tensor simaai::neat::from_gst_sample (GstSample * sample)

Definition at line 18 of file TensorAdapters.h.

graph_run_to_json()

std::string simaai::neat::graph_run_to_json (const Run & run, const GraphRunExportOptions & opt={}, std::string * err=nullptr)

inline

Compatibility wrapper for existing graph-run JSON call sites.

Definition at line 64 of file RunExport.h.

graph_run_to_json()

std::string simaai::neat::graph_run_to_json (const Run & run, const MeasureReport & report, const GraphRunExportOptions & opt={}, std::string * err=nullptr)

inline

Compatibility wrapper for measured graph-run JSON call sites.

Definition at line 70 of file RunExport.h.

gst_init_once()

void simaai::neat::gst_init_once ()

Initialize GStreamer exactly once across the process.

Safe to call from multiple threads and multiple times — only the first call performs the actual gst_init(); subsequent calls return immediately.

Definition at line 22 of file GstInit.h.

gst_message_to_string()

std::string simaai::neat::gst_message_to_string (GstMessage * msg)

Format a single GStreamer bus message into a stable, one-line string.

Parameters

msg	The bus message to format. Must be non-null.

Returns

Human-readable summary suitable for logs and GraphReport.

Definition at line 43 of file GstBusWatch.h.

input_type_from_media_type()

PayloadType simaai::neat::input_type_from_media_type (std::string_view media_type)

inline

Legacy helper name kept as a source-compatible bridge.

Definition at line 45 of file PayloadType.h.

is_raw_video_format()

bool simaai::neat::is_raw_video_format (FormatTag tag)

inline

True iff tag names a raw (uncompressed) video format.

Definition at line 265 of file FormatSpec.h.

is_tensor_payload_format()

bool simaai::neat::is_tensor_payload_format (FormatTag tag)

inline

True iff tag names a tensor payload (MLA, BBOX, ARGMAX, dtype kinds, EVXX aliases).

Definition at line 272 of file FormatSpec.h.

is_tessellated_bf16_format()

bool simaai::neat::is_tessellated_bf16_format (const std::string & fmt)

True iff fmt names a BF16 tessellated tensor format string.

Definition at line 20 of file TessellatedTensor.h.

is_tessellated_int8_format()

bool simaai::neat::is_tessellated_int8_format (const std::string & fmt)

True iff fmt names an INT8 tessellated tensor format string.

Definition at line 18 of file TessellatedTensor.h.

make_bundle_sample()

Sample simaai::neat::make_bundle_sample (std::initializer_list< Sample > fields)

inline

Construct a Bundle-kind Sample whose payload is the given list of inner Samples.

Definition at line 852 of file GraphOptions.h.

make_cpu_external_storage()

std::shared_ptr< TensorBuffer > simaai::neat::make_cpu_external_storage (void * data, std::size_t size_bytes, std::shared_ptr< void > holder={}, bool read_only=true)

Wrap a foreign CPU pointer as a TensorBuffer without copying.

Lifetime of the underlying memory is tracked via the holder shared_ptr — pass a custom deleter or a refcounted handle that keeps the original allocation alive. Set read_only to indicate the framework should not write through this storage.

Definition at line 561 of file TensorCore.h.

make_cpu_owned_storage()

std::shared_ptr< TensorBuffer > simaai::neat::make_cpu_owned_storage (std::size_t size_bytes)

Allocate a CPU-owned heap buffer of size_bytes, initialized to zero.

Definition at line 553 of file TensorCore.h.

make_encoded_sample()

Sample simaai::neat::make_encoded_sample (std::vector< uint8_t > bytes, std::string caps_string, int64_t pts_ns=-1, int64_t dts_ns=-1, int64_t duration_ns=-1)

Build an encoded Sample from raw bytes and caps metadata.

The returned Sample owns bytes (moved in) and carries the supplied caps string plus optional PTS/DTS/duration timestamps. Pass -1 to leave a timestamp unset.

Parameters

bytes	Encoded payload (e.g., H.264 access unit or NAL).
caps_string	Caps string describing the codec/profile.
pts_ns	Presentation timestamp in nanoseconds, or -1 if unknown.
dts_ns	Decode timestamp in nanoseconds, or -1 if unknown.
duration_ns	Frame duration in nanoseconds, or -1 if unknown.

Returns

A Sample wrapping the encoded payload with the supplied caps.

Definition at line 47 of file EncodedSampleUtil.h.

make_tensor_sample()

Sample simaai::neat::make_tensor_sample (const std::string & port_name, simaai::neat::Tensor tensor)

inline

Construct a TensorSet-kind Sample wrapping a single Tensor for the named port.

Definition at line 827 of file GraphOptions.h.

media_type_from_payload_type()

std::string simaai::neat::media_type_from_payload_type (PayloadType payload_type)

inline

Convert a payload family to the canonical internal media type when one exists.

Definition at line 50 of file PayloadType.h.

normalize_caps_format_for_media()

std::string simaai::neat::normalize_caps_format_for_media (std::string media_type, std::string format)

inline

Normalize a caps-format string conditionally on its media type.

Tensor caps (application/vnd.simaai.tensor) are normalized to canonical EVXX_ tokens; other media types (raw video, encoded video) are returned unchanged.

Parameters

media_type	GStreamer-style media type (e.g., application/vnd.simaai.tensor).
format	Format string to potentially normalize.

Returns

Normalized format if applicable, otherwise format unchanged.

Definition at line 197 of file FormatSpec.h.

normalize_tensor_caps_format()

std::string simaai::neat::normalize_tensor_caps_format (std::string value)

inline

Normalize tensor format aliases to the canonical EVXX_* token.

Maps user-facing or legacy spellings (FP32/FLOAT32, BF16/BFLOAT16, INT8/16/32) onto the EV-side caps form. Unknown values are passed through unchanged.

Definition at line 167 of file FormatSpec.h.

parse_bbox_bytes()

std::vector< Box > simaai::neat::parse_bbox_bytes (const std::vector< uint8_t > & bytes, int img_w, int img_h, int expected_topk, bool strict)

Parse a packed BBOX byte payload into typed Box records.

Parameters

bytes	Raw bytes (the BBOX caps payload).
img_w	Original image width used to clamp/scale coordinates.
img_h	Original image height used to clamp/scale coordinates.
expected_topk	Number of detections expected (advisory; see strict).
strict	If true, mismatches against expected_topk throw; if false, parse best-effort.

Returns

Parsed detections (size may be less than expected_topk).

Definition at line 222 of file DetectionTypes.h.

payload_type_from_media_type()

PayloadType simaai::neat::payload_type_from_media_type (std::string_view media_type)

inline

Convert an internal/legacy caps media string to the public payload family.

Definition at line 31 of file PayloadType.h.

prewarm_runtime()

void simaai::neat::prewarm_runtime ()

Initialize NEAT's runtime substrate early.

This performs the same one-time GStreamer/plugin setup that the first Graph::build() would otherwise perform. Calling it during application startup/model loading can hide cold plugin-registry cost from the first latency-sensitive build.

Definition at line 42 of file runtime.h.

prewarm_runtime_async()

std::future< void > simaai::neat::prewarm_runtime_async ()

Start NEAT runtime prewarm on a background thread.

The returned future rethrows any initialization error from prewarm_runtime().

Definition at line 49 of file runtime.h.

read_detection_format()

std::string simaai::neat::read_detection_format (const simaai::neat::Tensor & tensor)

Read the detection-format token from a tensor, if present.

Checks tensor.semantic.detection->format first. Falls back to the legacy tensor.semantic.tess->format location for back-compat with un-migrated producers; the fallback is scheduled for removal once all producers tag via tag_detection_format.

Returns

The format token, or empty string if neither slot carries one.

Definition at line 327 of file DetectionTypes.h.

read_feature_format()

std::string simaai::neat::read_feature_format (const Tensor & tensor)

Definition at line 73 of file FeatureTypes.h.

require_element()

void simaai::neat::require_element (const char * factory, const char * context)

Throw a NeatError if factory is not registered.

Parameters

factory	GStreamer element factory name (e.g., "h264parse").
context	Caller context interpolated into the error message (typically the Node kind).

Definition at line 33 of file GstHelpers.h.

require_single_tensor()

Tensor & simaai::neat::require_single_tensor (Sample & sample, const char * where=nullptr)

Returns the sample's single Tensor; throws if the sample carries 0 or >1 tensors.

Definition at line 875 of file GraphOptions.h.

require_single_tensor()

const Tensor & simaai::neat::require_single_tensor (const Sample & sample, const char * where=nullptr)

Const variant of require_single_tensor.

Definition at line 877 of file GraphOptions.h.

require_tensordecoder()

void simaai::neat::require_tensordecoder (const char * context)

Throw a NeatError if SiMa's tensordecoder element is not available.

Parameters

context

Caller context for the error message.

Definition at line 39 of file GstHelpers.h.

resolve_input_media_type()

std::string simaai::neat::resolve_input_media_type (const InputOptions & opt)

inline

Resolve the internal caps media type requested by public InputOptions.

Definition at line 130 of file Input.h.

run_to_json()

std::string simaai::neat::run_to_json (const Run & run, const RunExportOptions & opt={}, std::string * err=nullptr)

Serialize a built Run to sima.neat.graph_run JSON. Returns empty string on error.

Definition at line 48 of file RunExport.h.

run_to_json()

std::string simaai::neat::run_to_json (const Run & run, const MeasureReport & report, const RunExportOptions & opt={}, std::string * err=nullptr)

Serialize a built Run plus a measurement-window report to graph JSON.

Definition at line 52 of file RunExport.h.

sample_from_tensors()

Sample simaai::neat::sample_from_tensors (const TensorList & tensors)

Construct a TensorSet sample from a list of tensors.

Definition at line 881 of file GraphOptions.h.

sample_has_tensor_list()

bool simaai::neat::sample_has_tensor_list (const Sample & sample)

inline

Returns true if the sample is a TensorSet with at least one tensor.

Definition at line 866 of file GraphOptions.h.

sample_is_multi_output()

bool simaai::neat::sample_is_multi_output (const Sample & sample)

inline

Returns true if the sample carries multiple outputs (Bundle, or TensorSet with size > 1).

Definition at line 860 of file GraphOptions.h.

sample_media_type()

std::string simaai::neat::sample_media_type (const Sample & sample)

inline

Return the internal media type corresponding to a Sample's payload family.

Definition at line 819 of file GraphOptions.h.

sample_payload_type()

PayloadType simaai::neat::sample_payload_type (const Sample & sample)

inline

Return a Sample's standardized payload family, falling back from legacy media_type.

Definition at line 787 of file GraphOptions.h.

sample_tensor_list()

TensorList & simaai::neat::sample_tensor_list (Sample & sample, const char * where=nullptr)

Mutable reference to the sample's tensor list. Throws NeatError if kind != TensorSet.

Definition at line 871 of file GraphOptions.h.

sample_tensor_list()

const TensorList & simaai::neat::sample_tensor_list (const Sample & sample, const char * where=nullptr)

Const reference to the sample's tensor list. Throws NeatError if kind != TensorSet.

Definition at line 873 of file GraphOptions.h.

save_graph_run_json()

bool simaai::neat::save_graph_run_json (const Run & run, const std::string & path, const GraphRunExportOptions & opt={}, std::string * err=nullptr)

inline

Compatibility wrapper for existing graph-run JSON call sites.

Definition at line 77 of file RunExport.h.

save_graph_run_json()

bool simaai::neat::save_graph_run_json (const Run & run, const MeasureReport & report, const std::string & path, const GraphRunExportOptions & opt={}, std::string * err=nullptr)

inline

Compatibility wrapper for measured graph-run JSON call sites.

Definition at line 83 of file RunExport.h.

save_run_json()

bool simaai::neat::save_run_json (const Run & run, const std::string & path, const RunExportOptions & opt={}, std::string * err=nullptr)

Atomically write run_to_json(run, opt) to path using path + ".tmp" then rename.

Definition at line 56 of file RunExport.h.

save_run_json()

bool simaai::neat::save_run_json (const Run & run, const MeasureReport & report, const std::string & path, const RunExportOptions & opt={}, std::string * err=nullptr)

Atomically write run_to_json(run, report, opt) to path.

Definition at line 60 of file RunExport.h.

superpoint_output_format_token()

const char * simaai::neat::superpoint_output_format_token (SuperPointOutputFormat format)

constexpr

Definition at line 62 of file SuperPointTypes.h.

superpoint_profile_token()

const char * simaai::neat::superpoint_profile_token (SuperPointProfile profile)

constexpr

Definition at line 45 of file SuperPointTypes.h.

tag_detection_format()

void simaai::neat::tag_detection_format (simaai::neat::Tensor & tensor, std::string format)

Tag a tensor as carrying a detection-decoder wire format.

Sets tensor.semantic.detection = DetectionSpec{.format = format} so downstream consumers (and decode_bbox_tensor, pyneat.decode_bbox) can recognize the payload via the type-honest DetectionSpec slot instead of overloading TessSpec::format.

Parameters

tensor	Tensor to mutate.
format	Wire-format token (e.g., "BBOX").

Definition at line 316 of file DetectionTypes.h.

tag_detection_format_in_sample()

void simaai::neat::tag_detection_format_in_sample (simaai::neat::Sample & sample)

Walk a Sample tree and tag every recognized BBOX-payload Tensor with semantic.detection.format = "BBOX".

Called as a runtime finalizer at every chokepoint where detection-stage outputs exit the framework into user code (e.g., Run::pull_tensors_strict, Postprocess). Normalises the type metadata so consumers can rely on the type-honest DetectionSpec slot instead of inspecting payload_tag, Sample::format, or the legacy TessSpec::format overload.

Definition at line 345 of file DetectionTypes.h.

tag_feature_format()

void simaai::neat::tag_feature_format (Tensor & tensor, std::string format)

Definition at line 72 of file FeatureTypes.h.

tag_feature_format_in_sample()

void simaai::neat::tag_feature_format_in_sample (Sample & sample)

Definition at line 76 of file FeatureTypes.h.

tensor_format_is_bf16_alias()

bool simaai::neat::tensor_format_is_bf16_alias (std::string value)

inline

True iff value names BF16 (any of BF16, BFLOAT16, EVXX_BF16, EVXX_BFLOAT16).

Definition at line 132 of file FormatSpec.h.

tensor_format_is_fp32_alias()

bool simaai::neat::tensor_format_is_fp32_alias (std::string value)

inline

True iff value names FP32 (FP32, FLOAT32, EVXX_FLOAT32).

Definition at line 138 of file FormatSpec.h.

tensor_format_is_int16_alias()

bool simaai::neat::tensor_format_is_int16_alias (std::string value)

inline

True iff value names INT16 (INT16, EVXX_INT16).

Definition at line 150 of file FormatSpec.h.

tensor_format_is_int32_alias()

bool simaai::neat::tensor_format_is_int32_alias (std::string value)

inline

True iff value names INT32 (INT32, EVXX_INT32).

Definition at line 156 of file FormatSpec.h.

tensor_format_is_int8_alias()

bool simaai::neat::tensor_format_is_int8_alias (std::string value)

inline

True iff value names INT8 (INT8, EVXX_INT8).

Definition at line 144 of file FormatSpec.h.

tensor_to_pcm_audio()

PcmAudio simaai::neat::tensor_to_pcm_audio (const Tensor & tensor)

Definition at line 15 of file TensorAudio.h.

tensors_from_sample()

TensorList simaai::neat::tensors_from_sample (const Sample & sample, bool require_nonempty=true)

Extract all tensors from any sample shape (Tensor, TensorSet, Bundle); flattens recursively.

Definition at line 879 of file GraphOptions.h.

throw_if_bus_error()

void simaai::neat::throw_if_bus_error (GstElement * pipeline, BusMessageFn on_message, void * user_data, BusErrorFn on_error, void * error_user_data)

Drain the bus and throw a NeatError if any error message is found.

Drains exactly like drain_bus but additionally invokes on_error and raises a NeatError if any GST_MESSAGE_ERROR is observed.

Parameters

pipeline	Pipeline whose bus to drain.
on_message	Callback invoked for every message (may be null).
user_data	Opaque pointer forwarded to on_message.
on_error	Callback invoked once for the error line before throwing (may be null).
error_user_data	Opaque pointer forwarded to on_error.

Definition at line 69 of file GstBusWatch.h.

to_string()

const char * simaai::neat::to_string (ValidationSeverity s)

inline

Stable string label for a ValidationSeverity (for reports/logs).

Definition at line 36 of file ValidationReport.h.

upper_copy_ascii()

std::string simaai::neat::upper_copy_ascii (std::string value)

inline

ASCII upper-case copy of value (does not touch non-ASCII bytes).

Definition at line 125 of file FormatSpec.h.

usb_camera_ids()

std::vector< std::string > simaai::neat::usb_camera_ids ()

Stable IDs of the connected USB (UVC) cameras, for UsbCameraInputOptions::camera_name.

IDs follow libcamera's uvcvideo handler: "<controller>-<ports:config.interface>-<vid:pid>", for example "/base/pci@7000000000-3.1:1.0-046d:08e5". An ID is stable across boots and plug order and changes when the camera moves to another USB port.

Definition at line 57 of file UsbCameraInput.h.

usb_camera_info()

UsbCameraInfo simaai::neat::usb_camera_info (const UsbCameraInputOptions & opt)

Validate opt, find its camera, check the mode and read one frame.

decoded_format is the layout the camera's JPEGs decode to: set it as a SimaDecode out_format. A model reads NV16 directly: set its preprocessing input format to match (PreprocessColorFormat::NV16). Pass the result to UsbCameraInput(opt, info) so the camera is not opened a second time.

Exceptions

std::invalid_argument

and std::runtime_error as UsbCameraInput does.

Definition at line 68 of file UsbCameraInput.h.

Variables

kDecodedBoxColumns

int64_t simaai::neat::kDecodedBoxColumns = 6

constexpr

Number of columns in a decoded-boxes tensor: x1, y1, x2, y2, score, class_id.

Definition at line 242 of file DetectionTypes.h.

kDecodedMaskHeight

int64_t simaai::neat::kDecodedMaskHeight = 160

constexpr

Height of each BoxDecode instance mask.

Definition at line 131 of file DetectionTypes.h.

kDecodedMaskWidth

int64_t simaai::neat::kDecodedMaskWidth = 160

constexpr

Width of each BoxDecode instance mask.

Definition at line 129 of file DetectionTypes.h.

kDecodedPoseColumns

int64_t simaai::neat::kDecodedPoseColumns = 3

constexpr

Number of values per keypoint: x, y, visibility.

Definition at line 127 of file DetectionTypes.h.

kDecodedPoseKeypoints

int64_t simaai::neat::kDecodedPoseKeypoints = 17

constexpr

Number of keypoints emitted by the BoxDecode pose wire format.

Definition at line 125 of file DetectionTypes.h.

kDetectionFormatBbox

char simaai::neat::kDetectionFormatBbox[] = "BBOX"

constexpr

Canonical detection-format token for BoxDecode bounding-box payloads.

Definition at line 64 of file DetectionTypes.h.

kDetectionFormatBboxObbV1

char simaai::neat::kDetectionFormatBboxObbV1[] = "BBOX_OBB_V1"

constexpr

Versioned, little-endian oriented-box carrier; not an axis-aligned BBOX payload.

Definition at line 73 of file DetectionTypes.h.

kDetectionFormatBboxPose

char simaai::neat::kDetectionFormatBboxPose[] = "BBOX_POSE"

constexpr

Canonical detection-format token for BoxDecode pose payloads.

Definition at line 66 of file DetectionTypes.h.

kDetectionFormatBboxSegmentation

char simaai::neat::kDetectionFormatBboxSegmentation[] = "BBOX_SEGMENTATION"

constexpr

Canonical detection-format token for BoxDecode instance-segmentation payloads.

Definition at line 68 of file DetectionTypes.h.

kDetectionFormatBboxSegmentationPose

char simaai::neat::kDetectionFormatBboxSegmentationPose[] = "BBOX_SEGMENTATION_POSE"

constexpr

Canonical detection-format token for BoxDecode payloads carrying masks AND keypoints.

Definition at line 70 of file DetectionTypes.h.

kFeatureFormatLegacyA65V0

char simaai::neat::kFeatureFormatLegacyA65V0[] = "FEATURE_POINTS_LEGACY_A65_V0"

constexpr

Definition at line 19 of file FeatureTypes.h.

kFeatureFormatPointsV1

char simaai::neat::kFeatureFormatPointsV1[] = "FEATURE_POINTS_V1"

constexpr

Definition at line 18 of file FeatureTypes.h.

kFeaturePointsMagicV1

std::uint32_t simaai::neat::kFeaturePointsMagicV1 = 0x31565046U

constexpr

Definition at line 20 of file FeatureTypes.h.

kInvalidRuntimeNodeId

RuntimeNodeId simaai::neat::kInvalidRuntimeNodeId = static_cast<RuntimeNodeId>(-1)

constexpr static

Definition at line 24 of file GraphMetrics.h.

kMultipartHeaderCaptureMaxBlockBytes

std::size_t simaai::neat::kMultipartHeaderCaptureMaxBlockBytes = 64U * 1024U

constexpr

Maximum total size, in bytes, of one multipart part header block.

Definition at line 24 of file MultipartJpegDemux.h.

kMultipartHeaderCaptureMaxHeaders

std::size_t simaai::neat::kMultipartHeaderCaptureMaxHeaders = 64U

constexpr

Maximum number of distinct header names one capture allowlist may select.

Definition at line 18 of file MultipartJpegDemux.h.

kMultipartHeaderCaptureMaxLineBytes

std::size_t simaai::neat::kMultipartHeaderCaptureMaxLineBytes = 8U * 1024U

constexpr

Maximum length, in bytes, of one header line inside a multipart part header block.

Definition at line 22 of file MultipartJpegDemux.h.

kMultipartHeaderCaptureMaxNameBytes

std::size_t simaai::neat::kMultipartHeaderCaptureMaxNameBytes = 128U

constexpr

Maximum length, in bytes, of a single selected header name.

Definition at line 20 of file MultipartJpegDemux.h.

---

The documentation for this namespace was generated from the following files:

• ValidationReport.h
• GstBusWatch.h
• GstHelpers.h
• GstInit.h
• Model.h
• runtime.h
• MultipartJpegDemux.h
• Input.h
• UsbCameraInput.h
• Cast.h
• ImageTransform.h
• PreprocMath.h
• SimaDecode.h
• SimaEncode.h
• BoxDecodeType.h
• DetectionTypes.h
• EncodedSampleUtil.h
• FeatureTypes.h
• FormatSpec.h
• GraphMetrics.h
• GraphOptions.h
• PayloadType.h
• Run.h
• RunExport.h
• SuperPointTypes.h
• TelemetryMonitor.h
• Tensor.h
• TensorAdapters.h
• TensorAudio.h
• TensorConversion.h
• TensorCore.h
• TessellatedTensor.h

---

Generated via doxygen2docusaurus 2.0.0 by Doxygen 1.15.0.