Struct DeepDistributedQP 

pub struct DeepDistributedQP { /* private fields */ }

DeepDistributedQP: Deep Learning-Aided Distributed Optimization.

DeepDistributedQP combines operator splitting methods with learned policies to efficiently solve large-scale quadratic programming problems in a distributed manner.

Implementations

impl DeepDistributedQP

pub fn new( learning_rate: f32, num_consensus_nodes: usize, max_iterations: usize, tolerance: f32, ) -> Self

Creates a new DeepDistributedQP optimizer with the given configuration.

pub fn for_large_scale() -> Self

Creates DeepDistributedQP with configuration optimized for large-scale problems.

pub fn for_portfolio_optimization() -> Self

Creates DeepDistributedQP with configuration optimized for portfolio optimization.

pub fn with_config(config: DeepDistributedQPConfig) -> Self

Creates DeepDistributedQP with custom configuration.

pub fn qp_solver_stats(&self) -> HashMap<String, (usize, f32, f32, bool)>

Returns statistics about the distributed QP solver.

pub fn cumulative_speedup(&self) -> f32

Returns the cumulative speedup achieved.

pub fn distributed_memory_usage(&self) -> usize

Returns memory usage of consensus nodes and policy networks.

impl DeepDistributedQP

pub fn num_workers(&self) -> usize

Get number of consensus workers/nodes

pub fn learning_rate(&self) -> f32

Get current learning rate

pub fn communication_rounds(&self) -> usize

Get estimated communication rounds

pub fn synchronization_overhead(&self) -> f32

Get synchronization overhead estimate

pub fn solve_qp( &mut self, problem_id: &str, p: &Tensor, q: &Tensor, a: Option<&Tensor>, b: Option<&Tensor>, g: Option<&Tensor>, h: Option<&Tensor>, ) -> Result<Tensor>

Solves a quadratic programming problem with explicit matrices.

pub fn set_policy_weights( &mut self, param_id: &str, weights: Vec<Tensor>, biases: Vec<Tensor>, ) -> Result<()>

Sets custom policy network weights.

pub fn train_policy( &mut self, param_id: &str, experience_data: &[(Tensor, f32)], ) -> Result<()>

Trains the policy network on collected experience.

Trait Implementations

impl Clone for DeepDistributedQP

fn clone(&self) -> DeepDistributedQP

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for DeepDistributedQP

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Optimizer for DeepDistributedQP

fn update(&mut self, parameter: &mut Tensor, gradient: &Tensor) -> Result<()>

Updates a parameter based on its gradient.

fn zero_grad(&mut self)

Clears all accumulated gradients.

fn step(&mut self)

Performs a single optimization step.

fn get_lr(&self) -> f32

Gets the current learning rate.

fn set_lr(&mut self, lr: f32)

Sets a new learning rate.

fn accumulate_grad( &mut self, parameter: &mut Tensor, grad: &Tensor, ) -> Result<(), TrustformersError>

Accumulates gradients for gradient accumulation.

fn apply_accumulated_grads( &mut self, accumulation_steps: usize, ) -> Result<(), TrustformersError>

Applies accumulated gradients after gradient accumulation.

impl StatefulOptimizer for DeepDistributedQP

type Config = DeepDistributedQPConfig

The configuration type for this optimizer.

type State = StateMemoryStats

The state type used by this optimizer.

fn config(&self) -> &Self::Config

Gets a reference to the optimizer’s configuration.

fn state(&self) -> &Self::State

Gets a reference to the optimizer’s internal state.

fn state_mut(&mut self) -> &mut Self::State

Gets a mutable reference to the optimizer’s internal state.

fn state_dict(&self) -> Result<HashMap<String, Tensor>>

Saves the optimizer state to a dictionary for checkpointing.

fn load_state_dict(&mut self, state: HashMap<String, Tensor>) -> Result<()>

Loads optimizer state from a dictionary during checkpoint restoration.

fn memory_usage(&self) -> StateMemoryStats

Gets memory usage statistics for this optimizer.

fn reset_state(&mut self)

Resets the optimizer state (useful for training restarts).

fn num_parameters(&self) -> usize

Returns the number of parameters being optimized.