Module core

Expand description

Core functionality Argmin Optimizaton toolbox core

This crate contains the core functionality of argmin. If you just want to run an optimization method, this is not what you are looking for. However, if you want to implement your own solver based on the argmin architecture, you should find all necessary tools here.

Re-exports§

pub use executor::*;

Modules§

executor: Executor
file: Output parameter vectors to file
macros: Macros
slog_logger: Loggers based on the slog crate

Structs§

ArgminCheckpoint: Checkpoint
ArgminIterData: The datastructure which is returned by the next_iter method of the Solver trait.
ArgminKV: A simple key-value storage
ArgminResult: Final struct returned by the run method of Executor.
ArgminSlogKV: This type is necessary in order to be able to implement slog::KV on ArgminKV
ArgminSlogLogger: A logger based on slog
Error: The Error type, a wrapper around a dynamic error type.
IterState: Maintains the state from iteration to iteration of a solver
MinimalNoOperator: Minimal No-op operator which does nothing, really.
NoOperator: Fake Operators for testing No-op operator with free choice of the types
Observer: Container for observers which acts just like a single Observer by implementing Observe on it.
OpWrapper: This wraps an operator and keeps track of how often the cost, gradient and Hessian have been computed and how often the modify function has been called. Usually, this is an implementation detail unless a solver is needed within another solver (such as a line search within a gradient descent method), then it may be necessary to wrap the operator in an OpWrapper.
WriteToFile: Write parameter vectors to file

Enums§

ArgminError: Argmin error type
CheckpointMode: Defines at which intervals a checkpoint is saved.
ObserverMode: This is used to indicate how often the observer will observe the status. Never deactivates the observer, Always and Every(i) will call the observer in every or every ith iteration, respectively. NewBest will call the observer only, if a new best solution is found.
TerminationReason: Indicates why the optimization algorithm stopped
WriteToFileSerializer: Different kinds of serializers

Traits§

ArgminAdd: Add a T to self
ArgminConj: Return the conjugate
ArgminDiv: (Pointwise) Divide a T by self
ArgminDot: Dot/scalar product of T and self
ArgminEye: Identity matrix
ArgminFloat: Trait alias to simplify common trait bounds
ArgminInv: Compute the inverse (T) of self
ArgminLineSearch: Defines a common interface for line search methods.
ArgminMinMax: Minimum and Maximum of type T
ArgminMul: (Pointwise) Multiply a T with self
ArgminNLCGBetaUpdate: Common interface for beta update methods (Nonlinear-CG)
ArgminNorm: Compute the l2-norm (U) of self
ArgminOp: This trait needs to be implemented for every operator/cost function.
ArgminRandom: Create a random number
ArgminScaledAdd: Add a T scaled by an U to self
ArgminScaledSub: Subtract a T scaled by an U from self
ArgminSub: Subtract a T from self
ArgminTranspose: Transposing a type
ArgminTrustRegion: Defines a common interface to methods which calculate approximate steps for trust region methods.
ArgminWeightedDot: Dot/scalar product of T and self weighted by W (p^TWv)
ArgminZero: Return param vector of all zeros (for now, this is a hack. It should be done better)
ArgminZeroLike: Zero for dynamically sized objects
Observe: Defines the interface every Observer needs to expose
Solver: Solver

Functions§

load_checkpoint: Load a checkpoint from disk