101template <ComputeData InputType = Kakshya::DataVariant, ComputeData OutputType = InputType>
116 [[nodiscard]] std::string
get_name()
const override
137 return !stored.empty()
148 "{}: input failed validation, passing it through unchanged",
189 if constexpr (std::is_same_v<InputType, OutputType>) {
199 if (channels.empty()) {
202 for (
const auto& channel : channels) {
203 if (channel.empty()) {
206 for (
double sample : channel) {
207 if (std::isnan(sample)) {
210 if (std::isinf(sample)) {
#define MF_WARN(comp, ctx,...)
#define MF_DESCRIBE(...)
Requests a generated member description for a struct at namespace scope.
ConfigRefs stored_configs(Supports< Configs... >) const
static NumericChannels extract_numeric_data(const T &compute_data)
extract numeric data from single-variant types
Types that need an associated SignalSourceContainer to extract data.
@ BufferProcessing
Buffer processing (Buffers::BufferManager, processing chains).
@ Yantra
DSP algorithms, computational units, matrix operations, Grammar.
@ TEMPORAL
Time-based patterns, onset detection.
@ CUSTOM
User-defined analysis types.
@ SPATIAL
Multi-dimensional geometric analysis.
@ SPECTRAL
Spectral energy (FFT-based).
@ RECURSIVE
Recursive/nested extraction.
@ PATTERN_BASED
Extract based on pattern recognition.
TransformationScope
Portion of the data a transformer processes.
@ SELECTIVE_BANDS
Transform specific frequency/spatial bands.
@ CONDITIONAL
Transform based on dynamic conditions.
@ FULL_DATA
Transform entire data set.
TransformationQuality
Quality level requested from a transformer.
@ HIGH_QUALITY
High-quality transformation, may be slower.
@ DRAFT
Fast, low-quality transformation for previews.
@ ADAPTIVE
Quality adapts based on available computational resources.
@ STANDARD
Balanced quality/performance for real-time use.
@ REFERENCE
Maximum quality, computational cost is secondary.
OperationType
Operation categories for organization and discovery.
@ IN_PLACE
Sort data in-place (modifies input).
@ CROSS_MODAL
Sort one modality by features of another.
@ STANDARD
Traditional comparison-based sorting.
@ PARALLEL
Parallel with other operations.
std::vector< std::reference_wrapper< const OperationConfig > > ConfigRefs
Borrowed views of the configs an operation holds.
TransformationType
Categories of transformation for discovery and organization.
@ GENERATIVE
AI/ML-driven or algorithmic generation-based transformations.
@ GRANULAR
Granular synthesis and micro-temporal transformations.
@ MATHEMATICAL
Mathematical transformations (polynomial mapping, matrix operations).
@ CONVOLUTION
Convolution-based transformations (impulse response, filters).
TransformationStrategy
How a transformer handles its input.
@ BUFFERED
Create transformed copy (preserves input).
@ LAZY
Lazy evaluation transformation (future: coroutines).
@ CHUNKED
Transform in chunks for efficient processing.
@ INCREMENTAL
Progressive transformation with intermediate results.
@ IN_PLACE
Transform data in-place (modifies input).
@ STREAMING
Stream-based transformation for large data.
@ FULL_DATA
Extract all available data.
@ TARGETED_REGIONS
Extract only specific regions.
std::shared_ptr< container_family_t< C > > get_container() const
The associated container of one family.
T data
The actual computation data.
std::vector< Kakshya::DataDimension > dimensions
Data dimensional structure.
Kakshya::DataModality modality
Data modality (audio, image, spectral, etc.).
Facets facets
Typed results that travel with the data, read by type.
std::unordered_map< std::string, std::any > metadata
Associated metadata.
Input/Output container for computation pipeline data flow with structure preservation.
OperationConfig()=default
The config structs an operation class accepts, declared as using supports.