Debian Science Project
Summary
Machine Learning
pacchetti di apprendimento automatico per Debian Science

Questo metapacchetto installa i pacchetti utili all'apprendimento automatico. I pacchetti inclusi vanno dai sistemi di inferenza (esperti) basati sulla conoscenza fino al software che implementa i metodi statistici avanzati che attualmente dominano il campo.

Description

For a better overview of the project's availability as a Debian package, each head row has a color code according to this scheme:

If you discover a project which looks like a good candidate for Debian Science to you, or if you have prepared an unofficial Debian package, please do not hesitate to send a description of that project to the Debian Science mailing list

Links to other tasks

Debian Science Machine Learning packages

Official Debian packages with high relevance

autoclass
classificazione o clustering automatici
Versions of package autoclass
ReleaseVersionArchitectures
bookworm3.3.6.dfsg.2-1amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
bullseye3.3.6.dfsg.1-2amd64,arm64,armhf,i386
sid3.3.6.dfsg.2-3amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
forky3.3.6.dfsg.2-3amd64,arm64,armhf,i386,ppc64el,riscv64,s390x
trixie3.3.6.dfsg.2-2amd64,arm64,armel,armhf,i386,ppc64el,riscv64,s390x
Debtags of package autoclass:
fieldmathematics
interfacecommandline
roleprogram
scopeutility
useorganizing
Popcon: 18 users (166 upd.)*
Versions and Archs
License: DFSG free
Git

AutoClass risolve il problema di rilevare in modo automatico classi all'interno di dati (talvolta chiamato clustering o apprendimento non supervisionato), in contrasto con la generazione di descrizioni di classi da esempi già classificati (chiamato apprendimento supervisionato). Mira a scoprire le classi "naturali" all'interno dei dati. AutoClass è applicabile ad osservazioni di quelle cose che possono essere descritte da un insieme di attributi, senza fare riferimento ad altre cose. I valori dei dati corrispondenti ad ogni attributo sono limitati ad essere o numeri o elementi di un insieme finito di simboli. Con dati numerici deve essere fornito un errore di misurazione.

gprolog
GNU Prolog compiler
Maintainer: Salvador Abreu
Versions of package gprolog
ReleaseVersionArchitectures
sid1.4.5.0-3amd64,i386
trixie1.4.5.0-3amd64,i386
bookworm1.4.5.0-3amd64,i386
bullseye1.4.5.0-3amd64,i386
Debtags of package gprolog:
develcompiler, interpreter, lang:prolog
interfacecommandline
roleprogram
scopeutility
suitegnu
works-withsoftware:source
Popcon: 37 users (153 upd.)*
Versions and Archs
License: DFSG free

GNU Prolog is a free Prolog compiler with constraint solving over finite domains (FD). GNU Prolog is largely compliant with the ISO standard and is part of the Prolog Commons initiative.

This package contains the compiler and runtime system for the ISO standard version of GNU Prolog, including the prototype modules implementation.

libfann-dev
file header e librerie di sviluppo per FANN
Maintainer: Christian Kastner
Versions of package libfann-dev
ReleaseVersionArchitectures
trixie2.2.0+ds-8amd64,arm64,armel,armhf,i386,ppc64el,riscv64,s390x
bullseye2.2.0+ds-6amd64,arm64,armhf,i386
sid2.2.0+ds-8amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
bookworm2.2.0+ds-8amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
forky2.2.0+ds-8amd64,arm64,armhf,i386,ppc64el,riscv64,s390x
Debtags of package libfann-dev:
devellang:c, library
roledevel-lib
Popcon: 12 users (166 upd.)*
Versions and Archs
License: DFSG free
Git

La libreria FANN (Fast Artificial Neural Network, veloce rete neurale artificiale) è una libreria per rete neurale libera e open source che implementa reti neurali artificiali multilivello in C con la gestione sia di reti completamente connesse sia di reti connesse in modo sparso. Sono gestite le esecuzioni multipiattaforma in virgola fissa e in virgola mobile. Include un'infrastruttura per gestire in modo semplice insiemi di dati di addestramento. È facile da usare, versatile, ben documentata e veloce.

Questo pacchetto contiene i file header e le librerie statiche che sono necessari per sviluppare applicazioni per libfann.

libga-dev
C++ Library of Genetic Algorithm Components
Versions of package libga-dev
ReleaseVersionArchitectures
trixie2.4.7-6amd64,arm64,armel,armhf,i386,ppc64el,riscv64,s390x
bookworm2.4.7-6amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
forky2.4.7-6amd64,arm64,armhf,i386,ppc64el,riscv64,s390x
sid2.4.7-6amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
bullseye2.4.7-4amd64,arm64,armhf,i386
Debtags of package libga-dev:
devellibrary
roledevel-lib
Popcon: 9 users (160 upd.)*
Versions and Archs
License: DFSG free
Git

GAlib contains a set of C++ genetic algorithm objects. The library includes tools for using genetic algorithms to do optimization in any C++ program using any representation and genetic operators. The documentation includes an extensive overview of how to implement a genetic algorithm as well as examples illustrating customizations to the GAlib classes.

This package contains the development files.

liblinear-dev
file header e librerie di sviluppo per LIBLINEAR
Versions of package liblinear-dev
ReleaseVersionArchitectures
sid2.3.0+dfsg-5amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
experimental2.43+dfsg-1amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
bookworm2.3.0+dfsg-5amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
bullseye2.3.0+dfsg-5amd64,arm64,armhf,i386
trixie2.3.0+dfsg-5amd64,arm64,armel,armhf,i386,ppc64el,riscv64,s390x
forky2.3.0+dfsg-5amd64,arm64,armhf,i386,ppc64el,riscv64,s390x
upstream2.49
Debtags of package liblinear-dev:
devellang:c, lang:c++, library
roledevel-lib
Popcon: 15 users (168 upd.)*
Newer upstream!
License: DFSG free
Git

LIBLINEAR è una libreria per classificatori lineari per apprendimento per applicazioni su larga scala. Gestisce SVM (Support Vector Machines, macchine a vettori di supporto) con perdita L2 e L1, regressioni logistiche, classificazioni multi-classe e anche macchine di programmazione lineare (SVM con regolarizzazione L1). La sua complessità computazionale scala linearmente con il numero di esempi di addestramento, e ciò la rende uno dei più veloci risolutori di SVM in circolazione.

Questo pacchetto contiene i file header e le librerie statiche.

libmlpack-dev
libreria C++ di apprendimento automatico intuitiva, veloce e scalabile (librerie di sviluppo)
Versions of package libmlpack-dev
ReleaseVersionArchitectures
sid4.6.2-1amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
bullseye3.4.2-1amd64,arm64,i386
forky4.6.2-1amd64,arm64,armhf,i386,ppc64el,riscv64,s390x
trixie4.6.0-1amd64,arm64,armel,armhf,i386,ppc64el,riscv64,s390x
Popcon: 0 users (0 upd.)*
Versions and Archs
License: DFSG free
Git

Questo pacchetto contiene i file di sviluppo della libreria mlpack.

Machine Learning Pack (mlpack) è una libreria C++ di apprendimento automatico intuitiva, veloce e scalabile, pensata per essere l'analogo di LAPACK per l'apprendimento automatico. Mira a implementare una vasta gamma di metodi di apprendimento automatico e a funzionare come "coltellino svizzero" per i ricercatori nel campo dell'apprendimento automatico.

libocas-dev
file header e librerie di sviluppo per LIBOCAS
Maintainer: Christian Kastner
Versions of package libocas-dev
ReleaseVersionArchitectures
bookworm0.97+dfsg-8amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
trixie0.97+dfsg-8amd64,arm64,armel,armhf,i386,ppc64el,riscv64,s390x
forky0.97+dfsg-8amd64,arm64,armhf,i386,ppc64el,riscv64,s390x
sid0.97+dfsg-8amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
bullseye0.97+dfsg-6amd64,arm64,armhf,i386
Debtags of package libocas-dev:
devellang:c, library
roledevel-lib
Popcon: 7 users (160 upd.)*
Versions and Archs
License: DFSG free
Git

Questa libreria implementa OCAS (Optimized Cutting Plane Algorithm) per addestrare classificatori SVM (Support Vector Machine) lineari a partire da dati su larga scala. Il lavoro di calcolo di OCAS scala in modo lineare con il numero di esempi di addestramento. È uno dei risolutori SVM più veloci disponibili per la risoluzione di SVM L2 regolarizzate multiclasse e lineari.

Questo pacchetto contiene i file header e le librerie statiche.

libsvm-dev
file header per LIBSVM
Versions of package libsvm-dev
ReleaseVersionArchitectures
sid3.24+ds-6.1amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
experimental3.25+ds-1~exp1.1amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
forky3.24+ds-6.1amd64,arm64,armhf,i386,ppc64el,riscv64,s390x
trixie3.24+ds-6.1amd64,arm64,armel,armhf,i386,ppc64el,riscv64,s390x
bookworm3.24+ds-6amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
bullseye3.24+ds-6amd64,arm64,armhf,i386
upstream3.36
Debtags of package libsvm-dev:
devellibrary
roledevel-lib
Popcon: 10 users (165 upd.)*
Newer upstream!
License: DFSG free
Git

LIBSVM, una libreria per apprendimento macchina, è un pacchetto facile da usare per gestire regressione e classificazione di vettori, e SVM a una classe. Gestisce la classificazione multiclasse, output delle probabilità e selezione dei parametri.

Questo pacchetto contiene i file header di sviluppo.

libvigraimpex-dev
file di sviluppo per la libreria C++ di visione artificiale
Versions of package libvigraimpex-dev
ReleaseVersionArchitectures
bullseye1.11.1+dfsg-8amd64,arm64,armhf,i386
sid1.12.2+dfsg-2amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
forky1.12.2+dfsg-2amd64,arm64,armhf,i386,ppc64el,riscv64,s390x
trixie1.12.2+dfsg-2amd64,arm64,armel,armhf,i386,ppc64el,riscv64,s390x
bookworm1.11.1+dfsg-11amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
Debtags of package libvigraimpex-dev:
devellang:c++, library
roledevel-lib
works-withimage, image:raster
Popcon: 8 users (164 upd.)*
Versions and Archs
License: DFSG free
Git

VIGRA (Vision with Generic Algorithms, visione con algoritmi generici) è una libreria per visione artificiale che pone particolare attenzione sugli algoritmi flessibili, perché gli algoritmi rappresentano la principale conoscenza pratica in questo campo. La libreria è stata di conseguenza creata usando programmazione generica come introdotta da Stepanov e Musser ed esemplificata nella Standard Template Library C++. Mediante la scrittura di pochi adattatori (iteratori e metodo di accesso per immagini) è possibile usare gli algoritmi di VIGRA nelle proprie strutture dati, all'interno del proprio ambiente.

Questo pacchetto contiene i file di sviluppo e gli header necessari per compilare programmi e pacchetti che usano VIGRA.

mcl
algoritmo Cluster di Markov
Versions of package mcl
ReleaseVersionArchitectures
bullseye14-137+ds-9amd64,arm64,armhf,i386
bookworm22-282+ds-2amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
trixie22-282+ds-4amd64,arm64,ppc64el,riscv64,s390x
forky22-282+ds-4amd64,arm64,ppc64el,riscv64,s390x
sid22-282+ds-4amd64,arm64,mips64el,ppc64el,riscv64,s390x
Debtags of package mcl:
fieldmathematics
roleprogram
Popcon: 26 users (178 upd.)*
Versions and Archs
License: DFSG free
Git

Il pacchetto MCL è un'implementazione dell'algoritmo MCL, fornisce strumenti per manipolare matrici sparse (le strutture dati essenziali dell'algoritmo MCL) e per eseguire esperimenti su cluster.

MCL viene attualmente utilizzato nelle scienze come la biologia (rilevamento di famiglie proteiche, genomica), informatica (cluster dei nodi in reti peer-to-peer) e linguistica (analisi del testo).

The package is enhanced by the following packages: zoem
Please cite: Stijn van Dongen and Cei Abreu-Goodger: Using MCL to extract clusters from networks. (PubMed,eprint) Methods Mol Biol. 804:281-95 (2012)
Registry entries: Bio.tools 
mrgingham
strumento per trovare scacchiere per procedure di calibrazione visiva
Versions of package mrgingham
ReleaseVersionArchitectures
bookworm1.22-1amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
trixie1.26-1amd64,arm64,armel,armhf,i386,ppc64el,riscv64,s390x
forky1.26-1amd64,arm64,armhf,i386,ppc64el,riscv64,s390x
sid1.26-1amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
Popcon: 2 users (3 upd.)*
Versions and Archs
License: DFSG free
Git

Data un'immagine osservata contenente una scacchiera o una griglia di cerchi, mrgingham localizza la scacchiera nell'immagine e calcola in modo preciso la posizione degli angoli della scacchiera (o i centri dei cerchi). Ciò è simile alle procedure in OpenCV, ma è più veloce e robusto.

Questo pacchetto fornisce gli strumenti per l'utente.

octave-ga
codice per ottimizzazione genetica per Octave
Versions of package octave-ga
ReleaseVersionArchitectures
bookworm0.10.3-2all
bullseye0.10.2-1all
sid0.10.4-1all
forky0.10.4-1all
trixie0.10.4-1all
Debtags of package octave-ga:
devellang:octave, library
roledevel-lib
Popcon: 0 users (0 upd.)*
Versions and Archs
License: DFSG free
Git

Questo pacchetto fornisce funzioni per lavorare con algoritmi genetici in Octave, un software di calcolo numerico. Fornisce la funzione ga(), che funziona in modo simile ad altre funzioni di ottimizzazione in Octave.

Questo pacchetto di componenti aggiuntivi di Octave fa parte del progetto Octave-Forge.

python3-amp
??? missing short description for package python3-amp :-(
Versions of package python3-amp
ReleaseVersionArchitectures
bookworm0.6.1-1amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
bullseye0.6.1-1amd64,arm64,armhf,i386
Popcon: 2 users (0 upd.)*
Versions and Archs
License: DFSG free
Git
python3-fann2
collegamenti Python 3 per FANN
Versions of package python3-fann2
ReleaseVersionArchitectures
trixie1.2.0+ds-4amd64,arm64,armel,armhf,i386,ppc64el,riscv64,s390x
sid1.2.0+ds-4amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
forky1.2.0+ds-4amd64,arm64,armhf,i386,ppc64el,riscv64,s390x
bullseye1.2.0+ds-2amd64,arm64,armhf,i386
bookworm1.2.0+ds-4amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
Popcon: 13 users (157 upd.)*
Versions and Archs
License: DFSG free
Git

La libreria FANN (Fast Artificial Neural Network, veloce rete neurale artificiale) è una libreria per rete neurale libera e open source che implementa reti neurali artificiali multilivello in C con la gestione sia di reti completamente connesse sia di reti connesse in modo sparso.

Questo pacchetto contiene i collegamenti Python 3 per FANN.

python3-genetic
algoritmi genetici in Python
Versions of package python3-genetic
ReleaseVersionArchitectures
trixie0.1.1b+git20170527.98255cb-4all
bookworm0.1.1b+git20170527.98255cb-3all
bullseye0.1.1b+git20170527.98255cb-2all
sid0.1.1b+git20170527.98255cb-4all
forky0.1.1b+git20170527.98255cb-4all
Popcon: 12 users (158 upd.)*
Versions and Archs
License: DFSG free
Git

python3-genetic fornisce algoritmi genetici in Python 3, come usati spesso nell'intelligenza artificiale. Dovrebbe essere in grado di risolvere qualsiasi problema che consista nel minimizzare funzioni.

In questo pacchetto sono contenute alcune demo che usano Genetic, incluso un programma impressionantemente semplice che fornisce una soluzione per il rinomato TSP (Travelling Salesman Problem, problema del commesso viaggiatore). Ci si assicuri di leggere demo/genetic_demo_2.py per l'elenco dei geni "magici" speciali che rendono Genetic veramente divertente e... vivo!

python3-keras
deep learning framework running on Theano or TensorFlow
Versions of package python3-keras
ReleaseVersionArchitectures
bullseye2.3.1+dfsg-3all
Popcon: 3 users (0 upd.)*
Versions and Archs
License: DFSG free
Git

Keras is a Python library for machine learning based on deep (multi- layered) artificial neural networks (DNN), which follows a minimalistic and modular design with a focus on fast experimentation.

Features of DNNs like neural layers, cost functions, optimizers, initialization schemes, activation functions and regularization schemes are available in Keras a standalone modules which can be plugged together as wanted to create sequence models or more complex architectures. Keras supports convolutions neural networks (CNN, used for image recognition resp. classification) and recurrent neural networks (RNN, suitable for sequence analysis like in natural language processing).

It runs as an abstraction layer on the top of Theano (math expression compiler) by default, which makes it possible to accelerate the computations by using (GP)GPU devices. Alternatively, Keras could run on Google's TensorFlow (not yet available in Debian).

python3-mdp
toolkit modulare per elaborazione dei dati
Versions of package python3-mdp
ReleaseVersionArchitectures
forky3.6-10all
sid3.6-10all
bullseye3.6-1.1all
bookworm3.6-2amd64,arm64,mips64el,ppc64el
trixie3.6-10all
Popcon: 24 users (161 upd.)*
Versions and Archs
License: DFSG free
Git

Infrastruttura Python per elaborazione di dati per creare software per complesse elaborazioni di dati combinando in pipe e reti algoritmi largamente usati di apprendimento macchina. Gli algoritmi implementati includono: PCA (Principal Component Analysis), ICA (Independent Component Analysis), SFA (Slow Feature Analysis), ISFA (Independent Slow Feature Analysis), GNG (Growing Neural Gas), analisi fattoriale, FDA (Fisher Discriminant Analysis) e classificatori gaussiani.

The package is enhanced by the following packages: python3-sklearn
python3-mlpy
pacchetto Python ad alte prestazioni per modellazione predittiva
Versions of package python3-mlpy
ReleaseVersionArchitectures
bullseye3.5.0+ds-1.2all
bookworm3.5.0+ds-2all
Popcon: 74 users (0 upd.)*
Versions and Archs
License: DFSG free
Git

mlpy fornisce procedure di alto livello che gestiscono, con poche righe di codice, la progettazione di DAP (Data Analysis Protocol) per la preelaborazione, il clustering, la classificazione predittiva e la selezione di variabili. Sono disponibili dei metodi per ordinare variabili ed assegnarvi pesi, per il ricampionamento di dati, la valutazione degli errori e il landscaping di esperimenti.

mlpy include: SVM (Support Vector Machine), KNN (K Nearest Neighbor), FDA, SRDA, PDA, DLDA (Fisher, Spectral Regression, Penalized, Diagonal Linear Discriminant Analysis) per la classificazione e l'assegnazione di pesi alle variabili, I-RELIEF, DWT e FSSun per l'assegnazione di pesi alle variabili, RFE (Recursive Feature Elimination) e RFS (Recursive Forward Selection) per ordinare le variabili, DWT, UWT, CWT (Discrete, Undecimated, Continuous Wavelet Transform), attribuzione KNN, DTW (Dynamic Time Warping), clustering gerarchici, k-medoid, metodi di ricampionamento, funzioni metriche, indicatori Canberra.

python3-opencv
collegamenti Python 3 per la libreria per visione artificiale
Versions of package python3-opencv
ReleaseVersionArchitectures
bullseye4.5.1+dfsg-5amd64,arm64,armhf,i386
bookworm4.6.0+dfsg-12amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
trixie4.10.0+dfsg-5amd64,arm64,armel,armhf,i386,ppc64el,riscv64,s390x
forky4.10.0+dfsg-5amd64,arm64,armhf,i386,ppc64el,riscv64,s390x
sid4.10.0+dfsg-5amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
upstream4.12.0
Popcon: 287 users (580 upd.)*
Newer upstream!
License: DFSG free
Git

Questo pacchetto contiene i collegamenti Python 3 per la libreria OpenCV (Open Computer Vision).

La libreria Open Computer Vision è una raccolta di algoritmi e codice d'esempio per vari problemi relativi alla visione artificiale. La libreria è compatibile con IPL (Image Processing Library di Intel) e, se disponibile, può usare IPP (Integrated Performance Primitives di Intel) per ottenere prestazioni migliori.

OpenCV fornisce tipi di dati e operatori portabili di basso livello e un insieme di funzionalità di alto livello per l'acquisizione video, l'elaborazione e l'analisi di immagini, analisi strutturale, analisi del movimento e inseguimento di oggetti, riconoscimento di oggetti, calibrazione di videocamere e ricostruzione 3D.

Please cite: Gary Bradski and Adrian Kaehler: Learning OpenCV: Computer Vision with the OpenCV Library (2008)
Registry entries: SciCrunch 
python3-sklearn
moduli Python per l'apprendimento automatico e il data mining - Python 3
Versions of package python3-sklearn
ReleaseVersionArchitectures
trixie1.4.2+dfsg-8all
forky1.4.2+dfsg-8all
sid1.4.2+dfsg-8all
bullseye0.23.2-5all
bookworm1.2.1+dfsg-1all
upstream1.7.2
Popcon: 313 users (298 upd.)*
Newer upstream!
License: DFSG free
Git

scikit-learn è una raccolta di moduli Python relativi ad apprendimento automatico/statistico e data mining. Una lista non esaustiva di funzionalità incluse:

  • modelli misti gaussiani,
  • apprendimento tramite varietà,
  • kNN,
  • SVM (tramite LIBSVM).

Questo pacchetto contiene la versione per Python 3.

The package is enhanced by the following packages: python3-sklearn-pandas
Registry entries: Bio.tools  SciCrunch 
python3-statsmodels
modulo Python 3 per la stima di modelli statistici
Versions of package python3-statsmodels
ReleaseVersionArchitectures
bullseye0.12.2-1all
forky0.14.5+dfsg-1all
sid0.14.5+dfsg-1all
bookworm0.13.5+dfsg-7all
trixie0.14.4+dfsg-1all
Popcon: 92 users (260 upd.)*
Versions and Archs
License: DFSG free
Git

Il modulo Python 3 statsmodels fornisce classi e funzioni per la stima di diverse categorie di modelli statistici. Tra questi attualmente sono inclusi: modelli di regressione lineare, OLS, GLS, WLS e GLS con errori AR(p), modelli lineari generalizzati per diverse famiglie di distribuzione ed M stimatori per modelli lineari robusti. È disponibile un'ampia lista di statistiche dei risultati per ciascun problema di stima.

Please cite: Skipper Seabold and Josef Perktold: Statsmodels: Econometric and statistical modeling with python (eprint) (2010)
python3-thinc
Practical Machine Learning for NLP in Python
Versions of package python3-thinc
ReleaseVersionArchitectures
sid9.1.1-1amd64,arm64,armel,armhf,i386,mips64el,riscv64,s390x
bookworm8.1.7-1amd64,arm64,armhf,i386,mips64el,s390x
forky9.1.1-1amd64,arm64,armhf,i386,riscv64,s390x
Popcon: 0 users (0 upd.)*
Versions and Archs
License: DFSG free
Git

Thinc is the machine learning library powering spaCy https://spacy.io. It features a battle-tested linear model designed for large sparse learning problems, and a flexible neural network model under development for spaCy v2.0 https://spacy.io/usage/v2.

Thinc is a practical toolkit for implementing models that follow the "Embed, encode, attend, predict" architecture. It's designed to be easy to install, efficient for CPU usage and optimised for NLP and deep learning with text – in particular, hierarchically structured input and variable-length sequences.

python3-torch
Tensors and Dynamic neural networks in Python (Python Interface)
Versions of package python3-torch
ReleaseVersionArchitectures
sid2.6.0+dfsg-9amd64,arm64,ppc64el,riscv64,s390x
trixie2.6.0+dfsg-7amd64,arm64,ppc64el,riscv64,s390x
bookworm1.13.1+dfsg-4amd64,arm64,ppc64el,s390x
bullseye1.7.1-7amd64,arm64,armhf
forky2.6.0+dfsg-9amd64,arm64,ppc64el,riscv64,s390x
upstream2.8.0
Popcon: 196 users (218 upd.)*
Newer upstream!
License: DFSG free
Git

PyTorch is a Python package that provides two high-level features:

(1) Tensor computation (like NumPy) with strong GPU acceleration (2) Deep neural networks built on a tape-based autograd system

You can reuse your favorite Python packages such as NumPy, SciPy and Cython to extend PyTorch when needed.

This is the CPU-only version of PyTorch (Python interface).

Please cite: Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga, Alban Desmaison, Andreas Kopf, Edward Yang, Zachary DeVito, Martin Raison, Alykhan Tejani, Sasank Chilamkurthy, Benoit Steiner, Lu Fang, Junjie Bai and Soumith Chintala:
Registry entries: SciCrunch 
python3-torch-sparse
PyTorch Extension Library of Optimized Autograd Sparse Matrix Operations
Versions of package python3-torch-sparse
ReleaseVersionArchitectures
forky0.6.18-3amd64,arm64,ppc64el,riscv64,s390x
sid0.6.18-3amd64,arm64,ppc64el,riscv64,s390x
trixie0.6.18-3amd64,arm64,ppc64el,riscv64,s390x
Popcon: 11 users (155 upd.)*
Versions and Archs
License: DFSG free
Git

This package consists of a small extension library of optimized sparse matrix operations with autograd support.

This package installs the library for Python 3.

python3-vigra
collegamento Python 3 per la libreria C++ di visione artificiale
Versions of package python3-vigra
ReleaseVersionArchitectures
bookworm1.11.1+dfsg-11amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
bullseye1.11.1+dfsg-8amd64,arm64,armhf,i386
Popcon: 7 users (0 upd.)*
Versions and Archs
License: DFSG free
Git

VIGRA (Vision with Generic Algorithms, visione con algoritmi generici) è una libreria per visione artificiale che pone particolare attenzione sugli algoritmi flessibili, perché gli algoritmi rappresentano la principale conoscenza pratica in questo campo. La libreria è stata di conseguenza creata usando programmazione generica come introdotta da Stepanov e Musser ed esemplificata nella Standard Template Library C++. Mediante la scrittura di pochi adattatori (iteratori e metodo di accesso per immagini) è possibile usare gli algoritmi di VIGRA nelle proprie strutture dati, all'interno del proprio ambiente.

Questo pacchetto esporta la funzionalità della libreria VIGRA in Python 3.

r-cran-amore
GNU R: un pacchetto per reti neurali più flessibile
Versions of package r-cran-amore
ReleaseVersionArchitectures
trixie0.2-16-2amd64,arm64,armel,armhf,i386,ppc64el,riscv64,s390x
bullseye0.2-16-1amd64,arm64,armhf,i386
bookworm0.2-16-2amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
forky0.2-16-2amd64,arm64,armhf,i386,ppc64el,riscv64,s390x
sid0.2-16-2amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
Popcon: 20 users (162 upd.)*
Versions and Archs
License: DFSG free
Git

Questo pacchetto è nato per fornire agli utenti R il robusto algoritmo TAO per reti neurali. È cresciuto e può essere interessante per gli utenti che vogliono implementare il proprio algoritmo di addestramento o per quelli le cui esigenze risiedono solamente nello "spazio utente".

r-cran-bayesm
pacchetto GNU R per inferenza bayesiana
Versions of package r-cran-bayesm
ReleaseVersionArchitectures
forky3.1-6+dfsg-1amd64,arm64,armhf,i386,ppc64el,riscv64,s390x
sid3.1-6+dfsg-1amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
bookworm3.1-5+dfsg-1amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
trixie3.1-6+dfsg-1amd64,arm64,armel,armhf,i386,ppc64el,riscv64,s390x
bullseye3.1-4+dfsg-1amd64,arm64,armhf,i386
Debtags of package r-cran-bayesm:
fieldmathematics, statistics
suitegnu
Popcon: 24 users (174 upd.)*
Versions and Archs
License: DFSG free
Git

Il pacchetto bayesm copre molti modelli importanti usati in applicazioni di marketing e micro-econometria. Il pacchetto include:

  • regressione bayesiana (variabili dipendenti mono- o multi-variate);
  • logit multinomiale (MNL) e probit multinomiale (MNP);
  • probit multivariato;
  • misture multivariate di normali;
  • modelli lineari gerarchici con probabilità a priori e covariate normali;
  • logit multinomiale gerarchico con misture di normali come probabilità a priori e covariate;
  • analisi bayesiana di dati per l'analisi congiunta basata sulla scelta;
  • trattamento bayesiano di modelli di variabili strumentali lineari;
  • analisi di dati ordinali multivariati di sondaggi con scale di valori eterogenee (come in Rossi et al, JASA (01)).

Per riferimenti ulteriori, consultare il libro "Bayesian Statistics and Marketing" degli autori: Allenby, McCulloch e Rossi.

r-cran-class
pacchetto GNU R per classificazione
Maintainer: Dirk Eddelbuettel
Versions of package r-cran-class
ReleaseVersionArchitectures
forky7.3-23-1amd64,arm64,armhf,i386,ppc64el,riscv64,s390x
bookworm7.3-21-1amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
trixie7.3-23-1amd64,arm64,armel,armhf,i386,ppc64el,riscv64,s390x
bullseye7.3-18-1amd64,arm64,armhf,i386
sid7.3-23-1amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
Debtags of package r-cran-class:
devellang:r
roleshared-lib
sciencecalculation, modelling
useanalysing
Popcon: 855 users (676 upd.)*
Versions and Archs
License: DFSG free
Git

Il pacchetto class fornisce funzioni e insiemi di dati per gestire il capitolo 12 sulle "classificazioni" del libro "Modern Applied Statistics with S" (quarta edizione) di W.N. Venables e B.D. Ripley. Il seguente URL fornisce più dettagli su questo libro: http://www.stats.ox.ac.uk/pub/MASS4

r-cran-cluster
pacchetto GNU R per analisi di cluster di Rousseeuw et al.
Maintainer: Dirk Eddelbuettel
Versions of package r-cran-cluster
ReleaseVersionArchitectures
sid2.1.8.1-1amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
bookworm2.1.4-1amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
bullseye2.1.1-1amd64,arm64,armhf,i386
trixie2.1.8.1-1amd64,arm64,armel,armhf,i386,ppc64el,riscv64,s390x
forky2.1.8.1-1amd64,arm64,armhf,i386,ppc64el,riscv64,s390x
Debtags of package r-cran-cluster:
devellang:r, library
fieldstatistics
roleapp-data
suitegnu
Popcon: 761 users (648 upd.)*
Versions and Archs
License: DFSG free
Git

Questo pacchetto fornisce funzioni e insiemi di dati per l'analisi di cluster, originariamente scritti da Peter Rousseeuw, Anja Struyf e Mia Hubert.

Questo pacchetto fa parte dell'insieme di pacchetti "raccomandati" da R Core e forniti con i rilasci originali dei sorgenti di R stesso.

r-cran-gbm
GNU R package providing Generalized Boosted Regression Models
Versions of package r-cran-gbm
ReleaseVersionArchitectures
sid2.2.2-1amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
bookworm2.1.8.1-1amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
bullseye2.1.8-1amd64,arm64,armhf,i386
Popcon: 43 users (9 upd.)*
Versions and Archs
License: DFSG free
Git

This package implements extensions to Freund and Schapire's AdaBoost algorithm and Friedman's gradient boosting machine. Includes regression methods for least squares, absolute loss, t-distribution loss, quantile regression, logistic, multinomial logistic, Poisson, Cox proportional hazards partial likelihood, AdaBoost exponential loss, Huberized hinge loss, and Learning to Rank measures (LambdaMart).

r-cran-mass
pacchetto GNU R per MASS di Venables e Ripley
Maintainer: Dirk Eddelbuettel
Versions of package r-cran-mass
ReleaseVersionArchitectures
forky7.3-65-1amd64,arm64,armhf,i386,ppc64el,riscv64,s390x
bullseye7.3-53.1-1amd64,arm64,armhf,i386
bookworm7.3-58.2-1amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
trixie7.3-65-1amd64,arm64,armel,armhf,i386,ppc64el,riscv64,s390x
sid7.3-65-1amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
Debtags of package r-cran-mass:
devellang:r
fieldstatistics
suitegnu
Popcon: 946 users (701 upd.)*
Versions and Archs
License: DFSG free
Git

Il pacchetto MASS fornisce funzioni e insiemi di dati per gestire quanto illustrato nel libro "Modern Applied Statistics with S" (4° edizione) di W.N. Venables e B.D. Ripley. L'URL seguente fornisce ulteriori dettagli sul libro: URL: http://www.stats.ox.ac.uk/pub/MASS4

The package is enhanced by the following packages: r-cran-pscl
r-cran-mcmcpack
funzioni R per stima di modelli con catene di Markov Monte Carlo
Versions of package r-cran-mcmcpack
ReleaseVersionArchitectures
bullseye1.5-0-1amd64,arm64,armhf,i386
sid1.7-1-1amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
forky1.7-1-1amd64,arm64,armhf,i386,ppc64el,riscv64,s390x
trixie1.7-1-1amd64,arm64,armel,armhf,i386,ppc64el,riscv64,s390x
bookworm1.6-3-1amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
Debtags of package r-cran-mcmcpack:
devellang:r, library
fieldstatistics
roleapp-data
suitegnu
Popcon: 21 users (174 upd.)*
Versions and Archs
License: DFSG free
Git

Questo è un insieme di funzioni per GNU R che implementa vari modelli statistici ed econometrici usando le stime delle catene di Markov Monte Carlo (MCMC) che permettono di "risolvere" modelli che altrimenti non sarebbero trattabili con altre tecniche tradizionali, in particolare i problemi di statistica bayesiana (dove una o più probabilità a priori sono usate come parte della procedura di stima, invece di assumere l'ignoranza rispetto alle stime dei punti "reali"), benché il metodo MCMC può anche essere usato per risolvere problemi di statistica di frequenze con probabilità a priori non informative. Le tecniche MCMC sono preferibili anche rispetto alle stime dirette quando ci sono dati mancanti.

Attualmente sono implementati diverse funzioni di inferenza ecologica (EI) (per stimare attributi o comportamenti a livello di individuo a partire da dati aggregati, come risultati elettorali o di censimenti), così come modelli per panel lineari tradizionali e dati da campioni trasversali, alcune funzionalità di visualizzazione per diagnostica di EI, modelli di teoria di risposta agli item a due item (o stima di punto ideale), analisi fattoriale a risposta metrica, ordinale e mista e modelli per regressione gaussiana (lineare) e di Poisson, regressione logistica (o logit) e modelli probit a risposta ordinale e binaria.

I pacchetti suggeriti (r-cran-bayesm, -eco e -mnp) contengono modelli aggiuntivi che possono anch'essi essere utili per coloro che sono interessati a questo pacchetto.

The package is enhanced by the following packages: r-cran-mcmc r-cran-mnp
r-cran-metrics
GNU R evaluation metrics for machine learning
Versions of package r-cran-metrics
ReleaseVersionArchitectures
forky0.1.4-2all
sid0.1.4-2all
trixie0.1.4-2all
bullseye0.1.4-2all
bookworm0.1.4-2all
Popcon: 5 users (3 upd.)*
Versions and Archs
License: DFSG free
Git

An implementation of evaluation metrics in R that are commonly used in supervised machine learning. It implements metrics for regression, time series, binary classification, classification, and information retrieval problems. It has zero dependencies and a consistent, simple interface for all functions.

r-cran-mlbench
problemi per benchmark di apprendimento macchina per GNU R
Versions of package r-cran-mlbench
ReleaseVersionArchitectures
trixie2.1-6-1amd64,arm64,armel,armhf,i386,ppc64el,riscv64,s390x
forky2.1-6-1amd64,arm64,armhf,i386,ppc64el,riscv64,s390x
sid2.1-6-1amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
bullseye2.1-3-1amd64,arm64,armhf,i386
bookworm2.1-3-1amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
Popcon: 170 users (265 upd.)*
Versions and Archs
License: DFSG free
Git

Questo pacchetto per GNU R fornisce una raccolta di problemi artificiali e reali per il benchmark dell'apprendimento macchina, inclusi, ad esempio, diversi insiemi di dati dal repository UCI.

r-cran-mlr
Machine learning in GNU R
Versions of package r-cran-mlr
ReleaseVersionArchitectures
bullseye2.18.0+dfsg-1amd64,arm64,armhf,i386
sid2.19.2+dfsg-1amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
forky2.19.2+dfsg-1amd64,arm64,armhf,i386,ppc64el,riscv64,s390x
bookworm2.19.1+dfsg-1amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
trixie2.19.2+dfsg-1amd64,arm64,armel,armhf,i386,ppc64el,riscv64,s390x
upstream2.19.3
Popcon: 88 users (249 upd.)*
Newer upstream!
License: DFSG free
Git

Interface to a large number of classification and regression techniques, including machine-readable parameter descriptions. There is also an experimental extension for survival analysis, clustering and general, example-specific cost-sensitive learning. Generic resampling, including cross-validation, bootstrapping and subsampling. Hyperparameter tuning with modern optimization techniques, for single- and multi-objective problems. Filter and wrapper methods for feature selection. Extension of basic learners with additional operations common in machine learning, also allowing for easy nested resampling. Most operations can be parallelized.

r-cran-mnp
pacchetto GNU R per fit di modelli MNP (multinomial probit)
Versions of package r-cran-mnp
ReleaseVersionArchitectures
bookworm3.1-3-1amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
bullseye3.1-1-1amd64,arm64,armhf,i386
trixie3.1-5-1amd64,arm64,armel,armhf,i386,ppc64el,riscv64,s390x
forky3.1-5-1amd64,arm64,armhf,i386,ppc64el,riscv64,s390x
sid3.1-5-1amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
Debtags of package r-cran-mnp:
devellang:r, library
fieldstatistics
roleapp-data
suitegnu
Popcon: 19 users (173 upd.)*
Versions and Archs
License: DFSG free
Git

MNP è un pacchetto R che fa il fit di modelli bayesiani MNP (multinomial probit) con metodo Monte Carlo per catene di Markov (MCMC). Insieme al modello standard multinomial probit, può anche fare il fit di modelli con diversi insiemi di scelte per ciascuna osservazione e ordinamento completo * parziale di tutte le alternative disponibili. La stima è basata sull'algoritmo efficiente per data augmentation dei dati marginali sviluppato da Imai e van Dyk (2004).

r-cran-msm
modelli multistato di Markov e di Markov nascosti in tempo continuo per GNU R
Versions of package r-cran-msm
ReleaseVersionArchitectures
sid1.8.2-1amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
forky1.8.2-1amd64,arm64,armhf,i386,ppc64el,riscv64,s390x
trixie1.8.2-1amd64,arm64,armel,armhf,i386,ppc64el,riscv64,s390x
bookworm1.7-1amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
bullseye1.6.8-1amd64,arm64,armhf,i386
Debtags of package r-cran-msm:
interfacecommandline
roleprogram
Popcon: 101 users (253 upd.)*
Versions and Archs
License: DFSG free
Git

Funzioni per il fitting di modelli generici multi-stato di Markov e di Markov nascosti in tempo continuo a dati longitudinali. Sia le probabilità di transizione tra gli stati markoviani sia il processo di output del modello di Markov nascosto possono essere modellati in termini di covariate. Sono gestiti svariati schemi di osservazione, inclusi processi osservati ad intervalli arbitrari, processi osservati completamente e stati censurati.

Please cite: Christopher H. Jackson: Multi-State Models for Panel Data: The msm Package for R. Journal of Statistical Software 38(8):1-29 (2011)
r-cran-tgp
modelli bayesiani di processi gaussiani ad albero per GNU R
Versions of package r-cran-tgp
ReleaseVersionArchitectures
sid2.4-23-1amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
bullseye2.4-17-1amd64,arm64,armhf,i386
bookworm2.4-21-1amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
trixie2.4-23-1amd64,arm64,armel,armhf,i386,ppc64el,riscv64,s390x
forky2.4-23-1amd64,arm64,armhf,i386,ppc64el,riscv64,s390x
Popcon: 93 users (248 upd.)*
Versions and Archs
License: DFSG free
Git

Regressione e disegni sperimentali bayesiani non stazionari, semiparametrici non lineari con processi gaussiani (GP) con salti a LLM (Limiting Linear Model). I casi speciali implementati includono anche i modelli lineari bayesiani, CART, modelli lineari ad albero, GP stazionari separabili e isotropici e modelli GP a indice singolo. Fornisce funzioni di tracciamento grafi a 1 e 2 dimensioni (con capacità di proiezione e suddivisione) e disegno di albero, progettati per la visualizzazione dell'output di tgp-class. Sono gestiti modelli di analisi della sensibilità e modelli multi-risoluzione. Sono fornite anche funzioni per disegno sperimentale sequenziale e campionamento adattivo, inclusi ALM, ALC e miglioramento atteso. Quest'ultimo gestisce ottimizzazioni senza derivate di funzioni black-box con rumore.

toulbar2
ottimizzazione combinatoriale esatta per modelli grafici
Versions of package toulbar2
ReleaseVersionArchitectures
sid1.2.1+dfsg-0.1amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
forky1.2.1+dfsg-0.1amd64,arm64,armhf,i386,ppc64el,riscv64,s390x
trixie1.2.1+dfsg-0.1amd64,arm64,armel,armhf,i386,ppc64el,riscv64,s390x
bookworm1.1.1+dfsg-1amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
bullseye1.1.1+dfsg-1amd64,arm64,armhf,i386
Popcon: 17 users (170 upd.)*
Versions and Archs
License: DFSG free
Git

Toulbar2 è uno strumento per ottimizzazione discreta esatta per modelli grafici come reti di funzioni di costo, campi casuali di Markov, problemi di soddisfazione di vincoli pesati e reti bayesiane.

weka
algoritmi di apprendimento automatico per lavori di data mining
Versions of package weka
ReleaseVersionArchitectures
bullseye3.6.14-2all
trixie3.6.14-4all
sid3.6.14-4all
forky3.6.14-4all
bookworm3.6.14-3all
upstream3-8-6
Debtags of package weka:
fieldstatistics
interfacecommandline, x11
roleprogram
sciencecalculation
scopeutility
useanalysing, calculating
works-withdb, text
x11application
Popcon: 39 users (171 upd.)*
Newer upstream!
License: DFSG free
Git

Weka è una collezione di algoritmi di apprendimento automatico (machine learning) scritto in Java che può essere usato sia dalla riga di comando sia dal proprio codice Java. Weka è ideale per lo sviluppo di nuovi schemi di apprendimento automatico.

Gli schemi implementati comprendono alberi di decisione indotti, regole di apprendimento, modelli generatori di alberi, support vector machines, regressioni localmente pesate, apprendimento basato sulle istanze, bagging, boosting e stacking. Include inoltre metodi di clustering e uno strumento di apprendimento di regole associative. Oltre agli schemi di apprendimento veri e propri, Weka contiene una grande varietà di strumenti che possono essere utilizzati per pre-processare insiemi di dati.

Questo pacchetto contiene i binari e gli esempi.

Official Debian packages with lower relevance

libdlib-dev
C++ toolkit for machine learning and computer vision - development
Versions of package libdlib-dev
ReleaseVersionArchitectures
bookworm19.24+dfsg-1amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
sid19.24.6+dfsg-1amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
forky19.24.6+dfsg-1amd64,arm64,armhf,i386,ppc64el,riscv64,s390x
trixie19.24.6+dfsg-1amd64,arm64,armel,armhf,i386,ppc64el,riscv64,s390x
bullseye19.10-3.1amd64,arm64,armhf,i386
upstream20.0
Popcon: 5 users (10 upd.)*
Newer upstream!
License: DFSG free
Git

Dlib is a general purpose cross-platform open source software library written in the C++ programming language. It now contains software components for dealing with networking, threads, graphical interfaces, complex data structures, linear algebra, statistical machine learning, image processing, data mining, XML and text parsing, numerical optimization, Bayesian networks, and numerous other tasks.

This package contains the development headers.

libdlpack-dev
Open In Memory Tensor Structure
Versions of package libdlpack-dev
ReleaseVersionArchitectures
forky1.0-1amd64,arm64,armhf,i386,ppc64el,riscv64,s390x
sid1.0-1amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
bookworm0.6-1amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
bullseye0.0~git20200217.3ec0443-2amd64,arm64,armhf,i386
trixie1.0-1amd64,arm64,armel,armhf,i386,ppc64el,riscv64,s390x
upstream1.1
Popcon: 0 users (0 upd.)*
Newer upstream!
License: DFSG free
Git

DLPack is an open in-memory tensor structure to for sharing tensor among frameworks. DLPack enables

  • Easier sharing of operators between deep learning frameworks.
  • Easier wrapping of vendor level operator implementations, allowing collaboration when introducing new devices/ops.
  • Quick swapping of backend implementations, like different version of BLAS
  • For final users, this could bring more operators, and possibility of mixing usage between frameworks.

DLPack do not intend to implement of Tensor and Ops, but instead use this as common bridge to reuse tensor and ops across frameworks.

libfannj-java
FannJ a Java binding to the Fast Artificial Neural Network (FANN) C library
Versions of package libfannj-java
ReleaseVersionArchitectures
forky0.7-1all
bullseye0.3-2all
bookworm0.7-1all
trixie0.7-1all
sid0.7-1all
Popcon: 0 users (0 upd.)*
Versions and Archs
License: DFSG free
Git

Use FannJ if you have an existing ANN from the FANN project (libfann2) that you would like to access from Java. There are several GUI tools that will help you create and train an ANN.

libfclib-dev
read and write problems from the Friction Contact Library (headers)
Versions of package libfclib-dev
ReleaseVersionArchitectures
forky3.1.0+dfsg-3amd64,arm64,armhf,i386,ppc64el,riscv64,s390x
bullseye3.1.0+dfsg-2amd64,arm64,armhf,i386
bookworm3.1.0+dfsg-2amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
trixie3.1.0+dfsg-3amd64,arm64,armel,armhf,i386,ppc64el,riscv64,s390x
sid3.1.0+dfsg-3amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
Popcon: 0 users (0 upd.)*
Versions and Archs
License: DFSG free
Git

fclib is an open source collection of Frictional Contact (FC) problems stored in a specific HDF5 format, and an open source light implementation of Input/Output functions in C Language to read and write problems.

The goal of this work is to set up a collection of 2D and 3D Frictional Contact (FC) problems in order to set up a list of benchmarks; provide a standard framework for testing available and new algorithms; and share common formulations of problems in order to exchange data.

Fclib is an open-source scientific software primarily targeted at modeling and simulating nonsmooth dynamical systems

This package includes the libfclib development headers.

libmrgingham-dev
strumento per trovare scacchiere per procedure di calibrazione visiva
Versions of package libmrgingham-dev
ReleaseVersionArchitectures
forky1.26-1amd64,arm64,armhf,i386,ppc64el,riscv64,s390x
trixie1.26-1amd64,arm64,armel,armhf,i386,ppc64el,riscv64,s390x
sid1.26-1amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
bookworm1.22-1amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
Popcon: 0 users (0 upd.)*
Versions and Archs
License: DFSG free
Git

Data un'immagine osservata contenente una scacchiera o una griglia di cerchi, mrgingham localizza la scacchiera nell'immagine e calcola in modo preciso la posizione degli angoli della scacchiera (o i centri dei cerchi). Ciò è simile alle procedure in OpenCV, ma è più veloce e robusto.

Questo pacchetto fornisce le librerie C++ di sviluppo.

libxgboost-predictor-java
Java implementation of XGBoost predictor for online prediction tasks
Versions of package libxgboost-predictor-java
ReleaseVersionArchitectures
trixie0.3.1+dfsg-2all
forky0.3.1+dfsg-3all
bookworm0.3.1+dfsg-2all
sid0.3.1+dfsg-3all
Popcon: 0 users (0 upd.)*
Versions and Archs
License: DFSG free
Git

XGBoost is an optimized distributed gradient boosting library designed to be highly efficient, flexible and portable. It implements machine learning algorithms under the Gradient Boosting framework. XGBoost provides a parallel tree boosting (also known as GBDT, GBM) that solve many data science problems in a fast and accurate way. The same code runs on major distributed environment (Kubernetes, Hadoop, SGE, MPI, Dask) and can solve problems beyond billions of examples.

libxsmm-dev
Library for matrix operations and deep learning primitives
Versions of package libxsmm-dev
ReleaseVersionArchitectures
trixie1.17-4amd64
sid1.17-4amd64
bookworm1.17-2amd64
Popcon: 0 users (0 upd.)*
Versions and Archs
License: DFSG free
Git

LIBXSMM is a library targeting Intel Architecture for specialized dense and sparse matrix operations, and deep learning primitives.

This package contains the tools, static libraries and header files.

python3-hdmedians
high-dimensional medians in Python3
Versions of package python3-hdmedians
ReleaseVersionArchitectures
bullseye0.14.1-1amd64,arm64,armhf,i386
sid0.14.2-6amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
forky0.14.2-6amd64,arm64,armhf,i386,ppc64el,riscv64,s390x
trixie0.14.2-6amd64,arm64,armel,armhf,i386,ppc64el,riscv64,s390x
bookworm0.14.2-5amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
Popcon: 19 users (51 upd.)*
Versions and Archs
License: DFSG free
Git

Various definitions for a high-dimensional median exist and this Python package provides a number of fast implementations of these definitions. Medians are extremely useful due to their high breakdown point (up to 50% contamination) and have a number of nice applications in machine learning, computer vision, and high-dimensional statistics.

This package currently has implementations of medoid and geometric median with support for missing data using NaN.

python3-imblearn
libreria che fornisce tecniche di ricampionamento
Versions of package python3-imblearn
ReleaseVersionArchitectures
trixie0.12.4-1all
bullseye0.7.0-6all
bookworm0.10.0-1all
sid0.12.4-1all
forky0.12.4-1all
upstream0.14.0
Popcon: 12 users (2 upd.)*
Newer upstream!
License: DFSG free
Git

imbalanced-learn è un pacchetto Python che offre diverse tecniche di ricampionamento comunemente usate negli insiemi di dati che mostrano forti squilibri tra le classi.

È compatibile con scikit-learn e fa parte dei progetti scikit-learn-contrib.

python3-liac-arff
library for reading and writing ARFF files in Python
Versions of package python3-liac-arff
ReleaseVersionArchitectures
bullseye2.5.0-1all
sid2.5.0-6all
trixie2.5.0-6all
bookworm2.5.0-3all
forky2.5.0-6all
Popcon: 0 users (0 upd.)*
Versions and Archs
License: DFSG free
Git

The liac-arff module implements functions to read and write ARFF files in Python. It was created in the Connectionist Artificial Intelligence Laboratory (LIAC), which takes place at the Federal University of Rio Grande do Sul (UFRGS), in Brazil.

ARFF (Attribute-Relation File Format) is an file format specially created for describing datasets which are used commonly for machine learning experiments and software. This file format was created to be used in WEKA, the best representative software for machine learning automated experiments.

python3-mrgingham
strumento per trovare scacchiere per procedure di calibrazione visiva
Versions of package python3-mrgingham
ReleaseVersionArchitectures
forky1.26-1amd64,arm64,armhf,i386,ppc64el,riscv64,s390x
sid1.26-1amd64,arm64,armel,armhf,i386,mips64el,ppc64el,riscv64,s390x
trixie1.26-1amd64,arm64,armel,armhf,i386,ppc64el,riscv64,s390x
bookworm1.22-1amd64,arm64,armel,armhf,i386,mips64el,mipsel,ppc64el,s390x
Popcon: 0 users (1 upd.)*
Versions and Archs
License: DFSG free
Git

Data un'immagine osservata contenente una scacchiera o una griglia di cerchi, mrgingham localizza la scacchiera nell'immagine e calcola in modo preciso la posizione degli angoli della scacchiera (o i centri dei cerchi). Ciò è simile alle procedure in OpenCV, ma è più veloce e robusto.

Questo pacchetto fornisce le interfacce Python.

science-numericalcomputation
pacchetti Debian Science per il calcolo numerico
Versions of package science-numericalcomputation
ReleaseVersionArchitectures
bookworm1.14.5all
trixie1.15all
bullseye1.14.2all
forky1.15all
sid1.15all
Debtags of package science-numericalcomputation:
devellang:lisp
rolemetapackage, shared-lib
Popcon: 17 users (168 upd.)*
Versions and Archs
License: DFSG free
Git

Questo pacchetto installa i pacchetti Debian Science utili per il calcolo numerico. Il pacchetto fornisce un sistema di calcolo orientato ai vettori e di visualizzazione per il calcolo scientifico e l'analisi dei dati. Questi pacchetti sono simili ai sistemi commerciali come Matlab e IDL.

science-statistics
pacchetti Debian Science per la statistica
Versions of package science-statistics
ReleaseVersionArchitectures
bullseye1.14.2all
bookworm1.14.5all
trixie1.15all
forky1.15all
sid1.15all
Debtags of package science-statistics:
rolemetapackage
suitedebian
Popcon: 18 users (165 upd.)*
Versions and Archs
License: DFSG free
Git

Questo metapacchetto fa parte del Debian Pure Blend "Debian Science" e installa pacchetti relativi alla statistica. Questa è un'attività generica che può essere utile per qualsiasi lavoro scientifico. Dipende da moltissimi pacchetti R, oltre che da alcuni altri strumenti che sono utili per fare statistiche. Inoltre è suggerita l'attività Debian Science per la matematica per installare, in modo opzionale, tutto il software relativo alla matematica.

science-typesetting
pacchetti Debian Science per la composizione tipografica
Versions of package science-typesetting
ReleaseVersionArchitectures
bookworm1.14.5all
trixie1.15all
forky1.15all
sid1.15all
bullseye1.14.2all
Debtags of package science-typesetting:
rolemetapackage
suitedebian
Popcon: 16 users (156 upd.)*
Versions and Archs
License: DFSG free
Git

Questo metapacchetto installa i pacchetti Debian Science relativi alla composizione tipografica. Chi installa questo pacchetto potrebbe essere interessato al debtag use::typesetting.

Packaging has started and developers might try the packaging code in VCS

ask
Adaptive Sampling Kit for big experimental spaces
Versions of package ask
ReleaseVersionArchitectures
VCS1.1.1-3all
Versions and Archs
License: GPL-2
Debian package not available
Git
Version: 1.1.1-3

Adaptive Sampling Kit (ASK) is a toolkit for sampling big experimental spaces. When the space is small, the response can be measured for every point in the space. When the space is large, doing an exhaustive measurement is either not possible in terms of execution time or simply not practical. ASK tries to find good approximations of the response by sampling only a small fraction of the space. ASK features multiple active learning algorithms to prioritize the exploration of the interesting parts of the experimental space.

caffe-cuda
Fast, open framework for Deep Learning (Meta)
Versions of package caffe-cuda
ReleaseVersionArchitectures
VCS1.0.0+git20180821.99bd997-4all
Versions and Archs
License: CAFFE_LICENSE_BSD-2-Clause
Debian package not available
Git
Version: 1.0.0+git20180821.99bd997-4

Caffe is a deep learning framework made with expression, speed, and modularity in mind. It is developed by the Berkeley AI Research Lab (BAIR) and community contributors.

This metapackage pulls CUDA version of caffe:

  • caffe-tools-cuda
  • libcaffe-cuda*
  • python3-caffe-cuda And suggests these packages:

  • libcaffe-cuda-dev

  • caffe-doc

Note, this CUDA version cannot co-exist with the CPU_ONLY version.

Please cite: Yangqing Jia, Evan Shelhamer, Jeff Donahue, Sergey Karayev, Jonathan Long, Ross Girshick, Sergio Guadarrama and Trevor Darrell: Caffe: Convolutional Architecture for Fast Feature Embedding. (eprint) arXiv preprint arXiv:1408.5093 (2014)
libmkldnn-dev
Intel Math Kernel Library for Deep Neural Networks (dev)
Versions of package libmkldnn-dev
ReleaseVersionArchitectures
VCS2.0~beta4-1all
Versions and Archs
License: Apache-2.0
Debian package not available
Git
Version: 2.0~beta4-1

Intel(R) Math Kernel Library for Deep Neural Networks (Intel(R) MKL-DNN) is an open source performance library for deep learning applications. The library accelerates deep learning applications and framework on Intel(R) architecture. Intel(R) MKL-DNN contains vectorized and threaded building blocks which you can use to implement deep neural networks (DNN) with C and C++ interfaces.

DNN functionality optimized for Intel architecture is also included in Intel(R) Math Kernel Library (Intel(R) MKL). API in this implementation is not compatible with Intel MKL-DNN and does not include certain new and experimental features.

This package contains the header files, and symbol links to the shared object.

libroot-math-mlp-dev
Multi layer perceptron extension for ROOT - development files
Versions of package libroot-math-mlp-dev
ReleaseVersionArchitectures
VCS6.07.02+dfsg-1all
Versions and Archs
License: free
Debian package not available
Git
Version: 6.07.02+dfsg-1

The ROOT system provides a set of OO frameworks with all the functionality needed to handle and analyze large amounts of data efficiently.

This package contains development files of the mlp plug-in for ROOT, provides a Multi Layer Perceptron Neural Network package for ROOT.

libroot-montecarlo-vmc-dev
Virtual Monte-Carlo library for ROOT - development files
Versions of package libroot-montecarlo-vmc-dev
ReleaseVersionArchitectures
VCS6.07.02+dfsg-1all
Versions and Archs
License: free
Debian package not available
Git
Version: 6.07.02+dfsg-1

The ROOT system provides a set of OO frameworks with all the functionality needed to handle and analyze large amounts of data efficiently.

This package contains development files of the Vmc library for ROOT.

libroot-tmva-dev
Toolkit for multivariate data analysis - development files
Versions of package libroot-tmva-dev
ReleaseVersionArchitectures
VCS6.07.02+dfsg-1all
Versions and Archs
License: free
Debian package not available
Git
Version: 6.07.02+dfsg-1

The ROOT system provides a set of OO frameworks with all the functionality needed to handle and analyze large amounts of data efficiently.

The Toolkit for Multivariate Analysis (TMVA) provides a ROOT-integrated environment for the parallel processing and evaluation of MVA techniques to discriminate signal from background samples. It presently includes (ranked by complexity):

  • Rectangular cut optimisation
  • Correlated likelihood estimator (PDE approach)
  • Multi-dimensional likelihood estimator (PDE - range-search approach)
  • Fisher (and Mahalanobis) discriminant
  • H-Matrix (chi-squared) estimator
  • Artificial Neural Network (two different implementations)
  • Boosted Decision Trees

The TMVA package includes an implementation for each of these discrimination techniques, their training and testing (performance evaluation). In addition all these methods can be tested in parallel, and hence their performance on a particular data set may easily be compared.

This package provides development files of TMVA package for ROOT.

libshogun-dev
Large Scale Machine Learning Toolbox
Versions of package libshogun-dev
ReleaseVersionArchitectures
VCS3.2.0-7.5all
Versions and Archs
License: free
Debian package not available
Git
Version: 3.2.0-7.5

SHOGUN - is a new machine learning toolbox with focus on large scale kernel methods and especially on Support Vector Machines (SVM) with focus to bioinformatics. It provides a generic SVM object interfacing to several different SVM implementations. Each of the SVMs can be combined with a variety of the many kernels implemented. It can deal with weighted linear combination of a number of sub-kernels, each of which not necessarily working on the same domain, where an optimal sub-kernel weighting can be learned using Multiple Kernel Learning. Apart from SVM 2-class classification and regression problems, a number of linear methods like Linear Discriminant Analysis (LDA), Linear Programming Machine (LPM), (Kernel) Perceptrons and also algorithms to train hidden markov models are implemented. The input feature-objects can be dense, sparse or strings and of type int/short/double/char and can be converted into different feature types. Chains of preprocessors (e.g. substracting the mean) can be attached to each feature object allowing for on-the-fly pre-processing.

SHOGUN comes in different flavours, a stand-a-lone version and also with interfaces to Matlab(tm), R, Octave, Readline and Python. This package includes the developer files required to create stand-a-lone executables.

python3-lasagne
deep learning library build on the top of Theano (Python3 modules)
Versions of package python3-lasagne
ReleaseVersionArchitectures
VCS0.1+git20200419.5d3c63c+ds-1all
Versions and Archs
License: Expat
Debian package not available
Git
Version: 0.1+git20200419.5d3c63c+ds-1

Lasagne is a Python library to build and train deep (multi-layered) artificial neural networks on the top of Theano (math expression compiler). In comparison to other abstraction layers for that like e.g. Keras, it abstracts Theano as little as possible.

Lasagne supports networks like Convolutional Neural Networks (CNN, mostly used for image recognition resp. classification) and the Long Short-Term Memory type (LSTM, a subtype of Recurrent Neural Networks, RNN).

This package contains the modules for Python 3.

root-system
metapackage to install all ROOT packages
Versions of package root-system
ReleaseVersionArchitectures
VCS6.07.02+dfsg-1all
Versions and Archs
License: free
Debian package not available
Git
Version: 6.07.02+dfsg-1

The ROOT system provides a set of OO frameworks with all the functionality needed to handle and analyze large amounts of data efficiently.

With the data defined as a set of objects, specialized storage methods can give direct access to the separate attributes of the selected objects, without having to touch the bulk of the data. Included are histogramming methods in 1, 2 and 3 dimensions, curve fitting, function evaluation, minimization, graphics and visualization classes to allow the easy creation of an analysis system that can query and process the data interactively or in batch mode.

The command language, the scripting (or macro) language, and the programming language are all C++, thanks to the built-in CINT C++ interpreter. This interpreter removes the time consuming compile/link cycle, allowing for fast prototyping of the macros, and providing a good environment to learn C++. If more performance is needed, the interactively developed macros can be compiled using a C++ compiler.

The system has been designed in such a way that it can query its databases in parallel on MPP machines or on clusters of workstations or high-end PCs. ROOT is an open system that can be dynamically extended by linking external libraries. This makes ROOT a premier platform on which to build data acquisition, simulation and data analysis systems.

This package is a metapackage to ensure the installation of all possible ROOT packages on a system.

scilab-ann
Scilab module for artificial neural networks
Versions of package scilab-ann
ReleaseVersionArchitectures
VCS0.4.2.4-2all
Versions and Archs
License: free
Debian package not available
Git
Version: 0.4.2.4-2

This module implements artificial neural networks capabilities into the Scilab language. Current features are:

  • Only layered feedforward networks are supported directly at the moment (for others use the "hooks" provided)
  • Unlimited number of layers
  • Unlimited number of neurons per each layer separately
  • User defined activation function (defaults to logistic)
  • User defined error function (defaults to SSE)
  • Algorithms implemented so far:
  • standard (vanilla) with or without bias, on-line or batch
  • momentum with or without bias, on-line or batch
  • SuperSAB with or without bias, on-line or batch
  • Conjugate gradients
  • Jacobian computation
  • Computation of result of multiplication between "vector" and Hessian
  • Some helper functions provided
spacy
Industrial-strength Natural Language Processing (NLP)
Versions of package spacy
ReleaseVersionArchitectures
VCS2.2.3-1all
Versions and Archs
License: MIT
Debian package not available
Git
Version: 2.2.3-1

spaCy is a library for advanced Natural Language Processing in Python and Cython. It’s built on the very latest research, and was designed from day one to be used in real products. spaCy comes with pre-trained statistical models and word vectors, and currently supports tokenization for 30+ languages. It features the fastest syntactic parser in the world, convolutional neural network models for tagging, parsing and named entity recognition and easy deep learning integration.

streamlit
fast way to build custom ML tools
Versions of package streamlit
ReleaseVersionArchitectures
VCS0.56.0-1all
Versions and Archs
License: Apache-2.0
Debian package not available
Git
Version: 0.56.0-1

Streamlit lets you create apps for your machine learning projects with deceptively simple Python scripts. It supports hot-reloading, so your app updates live as you edit and save your file. No need to mess with HTTP requests, HTML, JavaScript, etc. All you need is your favorite editor and a browser.

Remark of Debian Science team: Needed for chime which is COVID-19 relevant.

Help for packaging is needed.

torch-core-free
Scientific Computing Framework For Luajit (Core Components)
Versions of package torch-core-free
ReleaseVersionArchitectures
VCS20171127all
Versions and Archs
License: BSD-3-Clause
Debian package not available
Git
Version: 20171127

Torch is a scientific computing framework with wide support for machine learning algorithms that puts GPUs first. It is easy to use and efficient, thanks to an easy and fast scripting language, LuaJIT, and an underlying C/CUDA implementation.

A summary of core features:

  • a powerful N-dimensional array
  • lots of routines for indexing, slicing, transposing, ...
  • amazing interface to C, via LuaJIT
  • linear algebra routines
  • neural network, and energy-based models
  • numeric optimization routines
  • Fast and efficient GPU support
  • Embeddable, with ports to iOS, Android and FPGA backends

The goal of Torch is to have maximum flexibility and speed in building your scientific algorithms while making the process extremely simple. Torch comes with a large ecosystem of community-driven packages in machine learning, computer vision, signal processing, parallel processing, image, video, audio and networking among others, and builds on top of the Lua community.

At the heart of Torch are the popular neural network and optimization libraries which are simple to use, while having maximum flexibility in implementing complex neural network topologies. You can build arbitrary graphs of neural networks, and parallelize them over CPUs and GPUs in an efficient manner.

This package is a metapackage, which pulls the following core and free modules for you: cwrap, paths, sys, xlua, torch7, nn, graph, nngraph, optim, sundown, dok, trepl, image.

Note that cutorch (CUDA backend for torch) and cunn (CUDA backend for neural network) are not present in this metapacakge - they will be shipped in the torch-core-contrib metapackage in the future.

Unofficial packages built by somebody else

python3-orange
Data mining framework
Responsible: Mitar
License: GPLv3

Orange is a component-based data mining software. It includes a range of data visualization, exploration, preprocessing and modeling techniques. It can be used through a nice and intuitive user interface or, for more advanced users, as a module for Python programming language.

No known packages available but some record of interest (WNPP bug)

flann - wnpp
Fast Library for Approximate Nearest Neighbors
License: BSD
Debian package not available
Language: C++

FLANN is a library for performing fast approximate nearest neighbor searches in high dimensional spaces. It contains a collection of algorithms we found to work best for nearest neighbor search and a system for automatically choosing the best algorithm and optimum parameters depending on the dataset.

pybrain - wnpp
Modular Machine Learning Library
License: BSD
Debian package not available
Language: Python

PyBrain is a modular machine learning library for Python. Its goal is to offer flexible, easy-to-use yet still powerful algorithms for Machine Learning Tasks and a variety of predefined environments to test and compare your algorithms.

PyBrain currently features algorithms for Supervised Learning, Unsupervised Learning, Reinforcment Learning and Black-box Optimization.

*Popularitycontest results: number of people who use this package regularly (number of people who upgraded this package recently) out of 268328