Package-level declarations
Types
Low level DenseTensorDataFactory factories.
Marker interface for tensor data backed by a contiguous FloatArray. Provides direct buffer access for performance-critical backends.
Marker interface for tensor data backed by a direct FloatBuffer. This enables zero-copy memory-mapped file access on the JVM.
Marker interface for tensor data backed by a contiguous IntArray.
A fundamental data structure interface that provides indexed access to elements.
Marker interface for tensor data backed by a MemorySegment. Backends can check for this interface to use MemorySegment-based SIMD operations (e.g. FloatVector.fromMemorySegment) instead of array-based ops.
Off-heap, 64-byte-aligned tensor data backed by a MemorySegment.
A TensorDataFactory that produces MemorySegmentTensorData tensors, keeping all data off-heap for SIMD-friendly access.
Memory-mapped tensor data implementation that provides zero-copy access to tensor data stored in files.
A handle to a memory-mapped file region that can provide tensor views.
Implementation of Q4_KTensorData backed by a packed byte array.
Tensor data interface for Q4_K quantized format.
Marker interface for Q4 quantized data backed by a MemorySegment. Enables DefaultCpuOpsJvm to dispatch to MemorySegment-based Q4 kernels.
Q4_0 quantized tensor data backed by a MemorySegment.
Implementation of Q8_0TensorData backed by a packed byte array.
Tensor data interface for Q8_0 quantized format.
Marker interface for Q8_0 quantized data backed by a MemorySegment. Enables DefaultCpuOpsJvm to dispatch to MemorySegment-based Q8 kernels.
Q8_0 quantized tensor data backed by a MemorySegment.
The fundamental data structure for tensor operations in the SKaiNET framework.
Interface for tensor data factory used by the DSL
Global registry for tensor data factories, enabling factory management for different precision types.
Ternary tensor data using 2-bit encoding compatible with TQ2_0 format.
Marker interface for tensor data containing ternary values {-1, 0, +1}.