sgd alternatives and similar packages
Based on the "Algorithms" category.
Alternatively, view sgd alternatives based on common mentions on social networks and blogs.
-
arithmoi
Number theory: primes, arithmetic functions, modular computations, special sequences -
imj-animation
Monorepo for a multi-player game engine, and game examples -
lca
Improves the known complexity of online lowest common ancestor search to O(log h) persistently, and without preprocessing -
search-algorithms
Haskell library containing common graph search algorithms -
integer-logarithms
Integer logarithms, originally split from arithmoi package -
treeviz
Haskell library for visualizing algorithmic decomposition of computations. -
incremental-sat-solver
Simple, Incremental SAT Solving as a Haskell Library -
nonlinear-optimization-ad
Wrapper of nonlinear-optimization package for using with ad and backprop packages -
infinite-search
An implementation of Martin Escardo's exhaustively searchable sets in Haskell. -
primesieve
A collection of packages related to math, algorithms and science, in Haskell. -
edit-distance-vector
Calculate edit scripts and distances between Vectors. -
graph-generators
A Haskell library for creating random Data.Graph instances using several pop -
edit-distance-linear
Levenshtein edit distance in linear memory (also turns out to be faster than C++) -
epanet-haskell
Call the EPANET toolkit via Haskell's Foreign Function Interface
Developer Ecosystem Survey 2022
* Code Quality Rankings and insights are calculated and provided by Lumnify.
They vary from L1 to L5 with "L5" being the highest.
Do you think we are missing an alternative of sgd or a related project?
Popular Comparisons
README
Haskell stochastic gradient descent library
Stochastic gradient descent (SGD) is a method for optimizing a global objective function defined as a sum of smaller, differentiable functions. In each iteration of SGD the gradient is calculated based on a subset of the training dataset. In Haskell, this process can be simply represented as a fold over a of subsequent dataset subsets (singleton elements in the extreme).
However, it can be beneficial to select the subsequent subsets randomly (e.g.,
shuffle the entire dataset before each pass). Moreover, the dataset can be
large enough to make it impractical to store it all in memory. Hence, the
sgd
library adopts a pipe-based
interface in which SGD takes the form of a process consuming dataset subsets
(the so-called mini-batches) and producing a stream of output parameter values.
The sgd
library implements several SGD variants (SGD with momentum, AdaDelta,
Adam) and handles heterogeneous parameter representations (vectors, maps, custom
records, etc.). It can be used in combination with automatic differentiation
libraries (ad,
backprop), which can be used to
automatically calculate the gradient of the objective function.
Look at the hackage repository for a library documentation.