Trait SpecializedGpuKernel

pub trait SpecializedGpuKernel: Send + Sync {
    // Required methods
    fn apply_holonomic_gate(
        &self,
        state: &mut dyn GpuBuffer,
        holonomy_matrix: &[Complex64],
        target_qubits: &[QubitId],
    ) -> QuantRS2Result<()>;
    fn apply_post_quantum_hash_gate(
        &self,
        state: &mut dyn GpuBuffer,
        hash_circuit: &[Complex64],
        compression_type: PostQuantumCompressionType,
    ) -> QuantRS2Result<()>;
    fn apply_quantum_ml_attention(
        &self,
        state: &mut dyn GpuBuffer,
        query_params: &[Complex64],
        key_params: &[Complex64],
        value_params: &[Complex64],
        num_heads: usize,
    ) -> QuantRS2Result<()>;
    fn apply_fused_gate_sequence(
        &self,
        state: &mut dyn GpuBuffer,
        gates: &[Box<dyn GateOp>],
    ) -> QuantRS2Result<()>;
    fn apply_tensor_contraction(
        &self,
        tensor_data: &mut dyn GpuBuffer,
        contraction_indices: &[usize],
        bond_dimension: usize,
    ) -> QuantRS2Result<()>;
}

Enhanced GPU kernel for specialized quantum operations

Required Methods

fn apply_holonomic_gate( &self, state: &mut dyn GpuBuffer, holonomy_matrix: &[Complex64], target_qubits: &[QubitId], ) -> QuantRS2Result<()>

Apply a holonomic gate with optimized GPU execution

fn apply_post_quantum_hash_gate( &self, state: &mut dyn GpuBuffer, hash_circuit: &[Complex64], compression_type: PostQuantumCompressionType, ) -> QuantRS2Result<()>

Apply post-quantum cryptographic hash gate

fn apply_quantum_ml_attention( &self, state: &mut dyn GpuBuffer, query_params: &[Complex64], key_params: &[Complex64], value_params: &[Complex64], num_heads: usize, ) -> QuantRS2Result<()>

Apply quantum ML attention mechanism

fn apply_fused_gate_sequence( &self, state: &mut dyn GpuBuffer, gates: &[Box<dyn GateOp>], ) -> QuantRS2Result<()>

Apply fused gate sequences for optimal performance

fn apply_tensor_contraction( &self, tensor_data: &mut dyn GpuBuffer, contraction_indices: &[usize], bond_dimension: usize, ) -> QuantRS2Result<()>

Apply tensor network contraction

Implementors