R-universe search: needs:coro

mlverse

torch:Tensors and Neural Networks with 'GPU' Acceleration

Provides functionality to define and train neural networks similar to 'PyTorch' by Paszke et al (2019) <doi:10.48550/arXiv.1912.01703> but written entirely in R using the 'libtorch' library. Also supports low-level tensor operations and 'GPU' acceleration.

Maintained by Daniel Falbel. Last updated 5 days ago.

autograd deep-learning torch cpp

521 stars 16.50 score 1.4k scripts 39 dependents

tidyverse

ellmer:Chat with Large Language Models

Chat with large language models from a range of providers including 'Claude' <https://claude.ai>, 'OpenAI' <https://chatgpt.com>, and more. Supports streaming, asynchronous calls, tool calling, and structured data extraction.

Maintained by Hadley Wickham. Last updated 3 days ago.

407 stars 12.94 score 98 scripts 8 dependents

michelnivard

gptstudio:Use Large Language Models Directly in your Development Environment

Large language models are readily accessible via API. This package lowers the barrier to use the API inside of your development environment. For more on the API, see <https://platform.openai.com/docs/introduction>.

Maintained by James Wade. Last updated 5 days ago.

chatgpt gpt-3 rstudio rstudio-addin

930 stars 10.85 score 43 scripts 1 dependents

mlverse

luz:Higher Level 'API' for 'torch'

A high level interface for 'torch' providing utilities to reduce the the amount of code needed for common tasks, abstract away torch details and make the same code work on both the 'CPU' and 'GPU'. It's flexible enough to support expressing a large range of models. It's heavily inspired by 'fastai' by Howard et al. (2020) <arXiv:2002.04688>, 'Keras' by Chollet et al. (2015) and 'PyTorch Lightning' by Falcon et al. (2019) <doi:10.5281/zenodo.3828935>.

Maintained by Daniel Falbel. Last updated 7 months ago.

89 stars 9.86 score 318 scripts 4 dependents

mlverse

torchvision:Models, Datasets and Transformations for Images

Provides access to datasets, models and preprocessing facilities for deep learning with images. Integrates seamlessly with the 'torch' package and it's 'API' borrows heavily from 'PyTorch' vision package.

Maintained by Daniel Falbel. Last updated 7 months ago.

65 stars 9.74 score 313 scripts 6 dependents

pln-team

PLNmodels:Poisson Lognormal Models

The Poisson-lognormal model and variants (Chiquet, Mariadassou and Robin, 2021 <doi:10.3389/fevo.2021.588292>) can be used for a variety of multivariate problems when count data are at play, including principal component analysis for count data, discriminant analysis, model-based clustering and network inference. Implements variational algorithms to fit such models accompanied with a set of functions for visualization and diagnostic.

Maintained by Julien Chiquet. Last updated 6 days ago.

count-data multivariate-analysis network-inference pca poisson-lognormal-model openblas cpp

55 stars 9.54 score 226 scripts

e-sensing

sits:Satellite Image Time Series Analysis for Earth Observation Data Cubes

An end-to-end toolkit for land use and land cover classification using big Earth observation data, based on machine learning methods applied to satellite image data cubes, as described in Simoes et al (2021) <doi:10.3390/rs13132428>. Builds regular data cubes from collections in AWS, Microsoft Planetary Computer, Brazil Data Cube, Copernicus Data Space Environment (CDSE), Digital Earth Africa, Digital Earth Australia, NASA HLS using the Spatio-temporal Asset Catalog (STAC) protocol (<https://stacspec.org/>) and the 'gdalcubes' R package developed by Appel and Pebesma (2019) <doi:10.3390/data4030092>. Supports visualization methods for images and time series and smoothing filters for dealing with noisy time series. Includes functions for quality assessment of training samples using self-organized maps as presented by Santos et al (2021) <doi:10.1016/j.isprsjprs.2021.04.014>. Includes methods to reduce training samples imbalance proposed by Chawla et al (2002) <doi:10.1613/jair.953>. Provides machine learning methods including support vector machines, random forests, extreme gradient boosting, multi-layer perceptrons, temporal convolutional neural networks proposed by Pelletier et al (2019) <doi:10.3390/rs11050523>, and temporal attention encoders by Garnot and Landrieu (2020) <doi:10.48550/arXiv.2007.00586>. Supports GPU processing of deep learning models using torch <https://torch.mlverse.org/>. Performs efficient classification of big Earth observation data cubes and includes functions for post-classification smoothing based on Bayesian inference as described by Camara et al (2024) <doi:10.3390/rs16234572>, and methods for active learning and uncertainty assessment. Supports region-based time series analysis using package supercells <https://jakubnowosad.com/supercells/>. Enables best practices for estimating area and assessing accuracy of land change as recommended by Olofsson et al (2014) <doi:10.1016/j.rse.2014.02.015>. Minimum recommended requirements: 16 GB RAM and 4 CPU dual-core.

Maintained by Gilberto Camara. Last updated 2 months ago.

big-earth-data cbers earth-observation eo-datacubes geospatial image-time-series land-cover-classification landsat planetary-computer r-spatial remote-sensing rspatial satellite-image-time-series satellite-imagery sentinel-2 stac-api stac-catalog cpp

494 stars 9.50 score 384 scripts

citoverse

cito:Building and Training Neural Networks

The 'cito' package provides a user-friendly interface for training and interpreting deep neural networks (DNN). 'cito' simplifies the fitting of DNNs by supporting the familiar formula syntax, hyperparameter tuning under cross-validation, and helps to detect and handle convergence problems. DNNs can be trained on CPU, GPU and MacOS GPUs. In addition, 'cito' has many downstream functionalities such as various explainable AI (xAI) metrics (e.g. variable importance, partial dependence plots, accumulated local effect plots, and effect estimates) to interpret trained DNNs. 'cito' optionally provides confidence intervals (and p-values) for all xAI metrics and predictions. At the same time, 'cito' is computationally efficient because it is based on the deep learning framework 'torch'. The 'torch' package is native to R, so no Python installation or other API is required for this package.

Maintained by Maximilian Pichler. Last updated 3 months ago.

machine-learning neural-network

42 stars 9.07 score 129 scripts 1 dependents

mlverse

tabnet:Fit 'TabNet' Models for Classification and Regression

Implements the 'TabNet' model by Sercan O. Arik et al. (2019) <doi:10.48550/arXiv.1908.07442> with 'Coherent Hierarchical Multi-label Classification Networks' by Giunchiglia et al. <doi:10.48550/arXiv.2010.10151> and provides a consistent interface for fitting and creating predictions. It's also fully compatible with the 'tidymodels' ecosystem.

Maintained by Christophe Regouby. Last updated 19 hours ago.

tabnet

109 stars 9.05 score 65 scripts

simonpcouch

chores:A Collection of Large Language Model Assistants

Provides a collection of ergonomic large language model assistants designed to help you complete repetitive, hard-to-automate tasks quickly. After selecting some code, press the keyboard shortcut you've chosen to trigger the package app, select an assistant, and watch your chore be carried out. While the package ships with a number of chore helpers for R package development, users can create custom helpers just by writing some instructions in a markdown file.

Maintained by Simon Couch. Last updated 1 months ago.

101 stars 8.44 score 6 scripts 1 dependents

mlr-org

mlr3torch:Deep Learning with 'mlr3'

Deep Learning library that extends the mlr3 framework by building upon the 'torch' package. It allows to conveniently build, train, and evaluate deep learning models without having to worry about low level details. Custom architectures can be created using the graph language defined in 'mlr3pipelines'.

Maintained by Sebastian Fischer. Last updated 2 months ago.

data-science deep-learning machine-learning mlr3 torch

42 stars 7.63 score 78 scripts

tidymodels

brulee:High-Level Modeling Functions with 'torch'

Provides high-level modeling functions to define and train models using the 'torch' R package. Models include linear, logistic, and multinomial regression as well as multilayer perceptrons.

Maintained by Max Kuhn. Last updated 11 days ago.

69 stars 7.56 score 214 scripts

bips-hb

innsight:Get the Insights of Your Neural Network

Interpretation methods for analyzing the behavior and individual predictions of modern neural networks in a three-step procedure: Converting the model, running the interpretation method, and visualizing the results. Implemented methods are, e.g., 'Connection Weights' described by Olden et al. (2004) <doi:10.1016/j.ecolmodel.2004.03.013>, layer-wise relevance propagation ('LRP') described by Bach et al. (2015) <doi:10.1371/journal.pone.0130140>, deep learning important features ('DeepLIFT') described by Shrikumar et al. (2017) <doi:10.48550/arXiv.1704.02685> and gradient-based methods like 'SmoothGrad' described by Smilkov et al. (2017) <doi:10.48550/arXiv.1706.03825>, 'Gradient x Input' or 'Vanilla Gradient'. Details can be found in the accompanying scientific paper: Koenen & Wright (2024, Journal of Statistical Software, <doi:10.18637/jss.v111.i08>).

Maintained by Niklas Koenen. Last updated 1 days ago.

30 stars 7.31 score 57 scripts

ashbythorpe

selenider:Concise, Lazy and Reliable Wrapper for 'chromote' and 'selenium'

A user-friendly wrapper for web automation, using either 'chromote' or 'selenium'. Provides a simple and consistent API to make web scraping and testing scripts easy to write and understand. Elements are lazy, and automatically wait for the website to be valid, resulting in reliable and reproducible code, with no visible impact on the experience of the programmer.

Maintained by Ashby Thorpe. Last updated 3 months ago.

web-scraping

39 stars 7.21 score 23 scripts

maxwell-geospatial

geodl:Geospatial Semantic Segmentation with Torch and Terra

Provides tools for semantic segmentation of geospatial data using convolutional neural network-based deep learning. Utility functions allow for creating masks, image chips, data frames listing image chips in a directory, and DataSets for use within DataLoaders. Additional functions are provided to serve as checks during the data preparation and training process. A UNet architecture can be defined with 4 blocks in the encoder, a bottleneck block, and 4 blocks in the decoder. The UNet can accept a variable number of input channels, and the user can define the number of feature maps produced in each encoder and decoder block and the bottleneck. Users can also choose to (1) replace all rectified linear unit (ReLU) activation functions with leaky ReLU or swish, (2) implement attention gates along the skip connections, (3) implement squeeze and excitation modules within the encoder blocks, (4) add residual connections within all blocks, (5) replace the bottleneck with a modified atrous spatial pyramid pooling (ASPP) module, and/or (6) implement deep supervision using predictions generated at each stage in the decoder. A unified focal loss framework is implemented after Yeung et al. (2022) <https://doi.org/10.1016/j.compmedimag.2021.102026>. We have also implemented assessment metrics using the 'luz' package including F1-score, recall, and precision. Trained models can be used to predict to spatial data without the need to generate chips from larger spatial extents. Functions are available for performing accuracy assessment. The package relies on 'torch' for implementing deep learning, which does not require the installation of a 'Python' environment. Raster geospatial data are handled with 'terra'. Models can be trained using a Compute Unified Device Architecture (CUDA)-enabled graphics processing unit (GPU); however, multi-GPU training is not supported by 'torch' in 'R'.

Maintained by Aaron Maxwell. Last updated 8 months ago.

12 stars 6.86 score 20 scripts

simonpcouch

gander:High Performance, Low Friction Large Language Model Chat

Introduces a 'Copilot'-like completion experience, but it knows how to talk to the objects in your R environment. 'ellmer' chats are integrated directly into your 'RStudio' and 'Positron' sessions, automatically incorporating relevant context from surrounding lines of code and your global environment (like data frame columns and types). Open the package dialog box with a keyboard shortcut, type your request, and the assistant will stream its response directly into your documents.

Maintained by Simon Couch. Last updated 1 months ago.

76 stars 6.54 score 1 scripts

duct317

scDHA:Single-Cell Decomposition using Hierarchical Autoencoder

Provides a fast and accurate pipeline for single-cell analyses. The 'scDHA' software package can perform clustering, dimension reduction and visualization, classification, and time-trajectory inference on single-cell data (Tran et.al. (2021) <DOI:10.1038/s41467-021-21312-2>).

Maintained by Ha Nguyen. Last updated 12 months ago.

cpp

40 stars 6.38 score 20 scripts 2 dependents

mlverse

torchdatasets:Ready to Use Extra Datasets for Torch

Provides datasets in a format that can be easily consumed by torch 'dataloaders'. Handles data downloading from multiple sources, caching and pre-processing so users can focus only on their model implementations.

Maintained by Daniel Falbel. Last updated 20 days ago.

15 stars 5.65 score 99 scripts

ericdunipace

causalOT:Optimal Transport Weights for Causal Inference

Uses optimal transport distances to find probabilistic matching estimators for causal inference. These methods are described in Dunipace, Eric (2021) <arXiv:2109.01991>. The package will build the weights, estimate treatment effects, and calculate confidence intervals via the methods described in the paper. The package also supports several other methods as described in the help files.

Maintained by Eric Dunipace. Last updated 8 months ago.

cpp

6 stars 5.38 score 20 scripts

dylanpieper

hellmer:Batch Processing for Chat Models

Batch processing framework for 'ellmer' chat models. Provides both sequential and parallel processing of chat interactions with features including tool calling and structured data extraction. Enables workflow management through progress tracking and recovery, automatic retry with backoff, and timeout handling. Additional quality-of-life features include verbosity control and sound notifications. Parallel processing is implemented via the 'future' framework. Includes methods for retrieving progress status, chat texts, and chat objects.

Maintained by Dylan Pieper. Last updated 10 days ago.

batch batch-processing ellmer llm

7 stars 5.32 score

e-sensing

torchopt:Advanced Optimizers for Torch

Optimizers for 'torch' deep learning library. These functions include recent results published in the literature and are not part of the optimizers offered in 'torch'. Prospective users should test these optimizers with their data, since performance depends on the specific problem being solved. The packages includes the following optimizers: (a) 'adabelief' by Zhuang et al (2020), <arXiv:2010.07468>; (b) 'adabound' by Luo et al.(2019), <arXiv:1902.09843>; (c) 'adahessian' by Yao et al.(2021) <arXiv:2006.00719>; (d) 'adamw' by Loshchilov & Hutter (2019), <arXiv:1711.05101>; (e) 'madgrad' by Defazio and Jelassi (2021), <arXiv:2101.11075>; (f) 'nadam' by Dozat (2019), <https://openreview.net/pdf/OM0jvwB8jIp57ZJjtNEZ.pdf>; (g) 'qhadam' by Ma and Yarats(2019), <arXiv:1810.06801>; (h) 'radam' by Liu et al. (2019), <arXiv:1908.03265>; (i) 'swats' by Shekar and Sochee (2018), <arXiv:1712.07628>; (j) 'yogi' by Zaheer et al.(2019), <https://papers.nips.cc/paper/8186-adaptive-methods-for-nonconvex-optimization>.

Maintained by Gilberto Camara. Last updated 2 years ago.

deep-learning numerical-optimization

26 stars 4.59 score 15 scripts

julien-hec

BKTR:Bayesian Kernelized Tensor Regression

Facilitates scalable spatiotemporally varying coefficient modelling with Bayesian kernelized tensor regression. The important features of this package are: (a) Enabling local temporal and spatial modeling of the relationship between the response variable and covariates. (b) Implementing the model described by Lei et al. (2023) <doi:10.48550/arXiv.2109.00046>. (c) Using a Bayesian Markov Chain Monte Carlo (MCMC) algorithm to sample from the posterior distribution of the model parameters. (d) Employing a tensor decomposition to reduce the number of estimated parameters. (e) Accelerating tensor operations and enabling graphics processing unit (GPU) acceleration with the 'torch' package.

Maintained by Julien Lanthier. Last updated 8 months ago.

2 stars 4.53 score 17 scripts

bioc

SCFA:SCFA: Subtyping via Consensus Factor Analysis

Subtyping via Consensus Factor Analysis (SCFA) can efficiently remove noisy signals from consistent molecular patterns in multi-omics data. SCFA first uses an autoencoder to select only important features and then repeatedly performs factor analysis to represent the data with different numbers of factors. Using these representations, it can reliably identify cancer subtypes and accurately predict risk scores of patients.

Maintained by Duc Tran. Last updated 5 months ago.

survival clustering classification

3 stars 4.48 score 2 scripts

neural-structured-additive-learning

deeptrafo:Fitting Deep Conditional Transformation Models

Allows for the specification of deep conditional transformation models (DCTMs) and ordinal neural network transformation models, as described in Baumann et al (2021) <doi:10.1007/978-3-030-86523-8_1> and Kook et al (2022) <doi:10.1016/j.patcog.2021.108263>. Extensions such as autoregressive DCTMs (Ruegamer et al, 2023, <doi:10.1007/s11222-023-10212-8>) and transformation ensembles (Kook et al, 2022, <doi:10.48550/arXiv.2205.12729>) are implemented. The software package is described in Kook et al (2024, <doi:10.18637/jss.v111.i10>).

Maintained by Lucas Kook. Last updated 2 months ago.

5 stars 4.44 score 11 scripts

simonpcouch

streamy:Inline Asynchronous Generator Results into Documents

Given a 'coro' asynchronous generator instance that produces text, write that text into a document selection in 'RStudio' and 'Positron'. This is particularly helpful for streaming large language model responses into the user's editor.

Maintained by Simon Couch. Last updated 2 months ago.

1 stars 4.43 score 3 dependents

opasche

EQRN:Extreme Quantile Regression Neural Networks for Risk Forecasting

This framework enables forecasting and extrapolating measures of conditional risk (e.g. of extreme or unprecedented events), including quantiles and exceedance probabilities, using extreme value statistics and flexible neural network architectures. It allows for capturing complex multivariate dependencies, including dependencies between observations, such as sequential dependence (time-series). The methodology was introduced in Pasche and Engelke (2024) <doi:10.1214/24-AOAS1907> (also available in preprint: Pasche and Engelke (2022) <doi:10.48550/arXiv.2208.07590>).

Maintained by Olivier C. Pasche. Last updated 10 days ago.

7 stars 4.24 score

barbaratarantino

SEMdeep:Structural Equation Modeling with Deep Neural Network and Machine Learning

Training and validation of a custom (or data-driven) Structural Equation Models using layer-wise Deep Neural Networks or node-wise Machine Learning algorithms, which extend the fitting procedures of the 'SEMgraph' R package <doi:10.32614/CRAN.package.SEMgraph>.

Maintained by Barbara Tarantino. Last updated 2 months ago.

4 stars 4.15 score

lsablica

circlus:Clustering and Simulation of Spherical Cauchy and PKBD Models

Provides tools for estimation and clustering of spherical data, seamlessly integrated with the 'flexmix' package. Includes the necessary M-step implementations for both Poisson Kernel-Based Distribution (PKBD) and spherical Cauchy distribution. Additionally, the package provides random number generators for PKBD and spherical Cauchy distribution. Methods are based on Golzy M., Markatou M. (2020) <doi:10.1080/10618600.2020.1740713>, Kato S., McCullagh P. (2020) <doi:10.3150/20-bej1222> and Sablica L., Hornik K., Leydold J. (2023) <doi:10.1214/23-ejs2149>.

Maintained by Lukas Sablica. Last updated 28 days ago.

openblas cpp openmp

4.04 score 1 scripts

wikihistories

wikkitidy:Tidy Analysis of Wikipedia

Access 'Wikipedia' through the several 'MediaWiki' APIs (<https://www.mediawiki.org/wiki/API>), as well as through the 'XTools' API (<https://www.mediawiki.org/wiki/XTools/API>). Ensure your API calls are correct, and receive results in tidy tibbles.

Maintained by Michael Falk. Last updated 2 months ago.

7 stars 4.02 score 2 scripts

mneunhoe

RGAN:Generative Adversarial Nets (GAN) in R

An easy way to get started with Generative Adversarial Nets (GAN) in R. The GAN algorithm was initially described by Goodfellow et al. 2014 <https://proceedings.neurips.cc/paper/2014/file/5ca3e9b122f61f8f06494c97b1afccf3-Paper.pdf>. A GAN can be used to learn the joint distribution of complex data by comparison. A GAN consists of two neural networks a Generator and a Discriminator, where the two neural networks play an adversarial minimax game. Built-in GAN models make the training of GANs in R possible in one line and make it easy to experiment with different design choices (e.g. different network architectures, value functions, optimizers). The built-in GAN models work with tabular data (e.g. to produce synthetic data) and image data. Methods to post-process the output of GAN models to enhance the quality of samples are available.

Maintained by Marcel Neunhoeffer. Last updated 2 years ago.

19 stars 3.98 score 7 scripts

mlverse

torchvisionlib:Additional Operators for Image Models

Implements additional operators for computer vision models, including operators necessary for image segmentation and object detection deep learning models.

Maintained by Daniel Falbel. Last updated 1 years ago.

cpp

9 stars 3.77 score 13 scripts

cran

glmnetr:Nested Cross Validation for the Relaxed Lasso and Other Machine Learning Models

Cross validation informed Relaxed LASSO, Artificial Neural Network (ANN), gradient boosting machine ('xgboost'), Random Forest ('RandomForestSRC'), Oblique Random Forest ('aorsf'), Recursive Partitioning ('RPART') or step wise regression models are fit. Cross validation leave out samples (leading to nested cross validation) or bootstrap out-of-bag samples are used to evaluate and compare performances between these models with results presented in tabular or graphical means. Calibration plots can also be generated, again based upon (outer nested) cross validation or bootstrap leave out (out of bag) samples. For some datasets, for example when the design matrix is not of full rank, 'glmnet' may have very long run times when fitting the relaxed lasso model, from our experience when fitting Cox models on data with many predictors and many patients, making it difficult to get solutions from either glmnet() or cv.glmnet(). This may be remedied by using the 'path=TRUE' option when calling glmnet() and cv.glmnet(). Within the glmnetr package the approach of path=TRUE is taken by default. When fitting not a relaxed lasso model but an elastic-net model, then the R-packages 'nestedcv' <https://cran.r-project.org/package=nestedcv>, 'glmnetSE' <https://cran.r-project.org/package=glmnetSE> or others may provide greater functionality when performing a nested CV. Use of the 'glmnetr' has many similarities to the 'glmnet' package and it is recommended that the user of 'glmnetr' also become familiar with the 'glmnet' package <https://cran.r-project.org/package=glmnet>, with the "An Introduction to 'glmnet'" and "The Relaxed Lasso" being especially useful in this regard.

Maintained by Walter K Kremers. Last updated 2 months ago.

3.60 score

neferkareii

shrinkGPR:Scalable Gaussian Process Regression with Hierarchical Shrinkage Priors

Efficient variational inference methods for fully Bayesian Gaussian Process Regression (GPR) models with hierarchical shrinkage priors, including the triple gamma prior for effective variable selection and covariance shrinkage in high-dimensional settings. The package leverages normalizing flows to approximate complex posterior distributions. For details on implementation, see Knaus (2025) <doi:10.48550/arXiv.2501.13173>.

Maintained by Peter Knaus. Last updated 2 months ago.

1 stars 3.48 score

frankiethull

kuzco:LLM image classification using ollama in R

This package is a designed to use local models for image classification. The prompts and functions are designed to take an input image and supply classification information as an output.

Maintained by Frank Hull. Last updated 2 months ago.

15 stars 3.48 score

vankesteren

tensorsem:Estimate structural equation models using computation graphs

Use lavaan code to create structural equation models, use torch to estimate them. This package provides the interface between lavaan and torch.

Maintained by Erik-Jan van Kesteren. Last updated 2 years ago.

computation-graph lavaan sem torch

52 stars 3.41 score 8 scripts

tadascience

valentine:Spread the Love for R Packages with Poetry

Uses large language models to create poems about R packages. Currently contains the roses() function to make "roses are red, ..." style poems and the prompt() function to only assemble the prompt without submitting it to the model.

Maintained by Romain François. Last updated 2 months ago.

ai chatgpt ellmer love poetry valentine valentine-day

1 stars 3.18 score 2 scripts

openvolley

ovml:Machine Learning Tools for Volleyball

Image and video machine learning tools, for application to volleyball analytics.

Maintained by Ben Raymond. Last updated 3 years ago.

23 stars 3.06 score 4 scripts

rdinnager

trampoline:Make Functions that Can Recurse Infinitely

Implements a trampoline algorithm for R that let's users write recursive functions that get around R's stack call limitations, enabling theoretically infinite recursion. The algorithm is based around generator function as implemented in the 'coro' package, and is based almost completely on the 'trampoline' module from Python <https://gitlab.com/ferreum/trampoline>.

Maintained by Russell Dinnage. Last updated 3 years ago.

21 stars 3.02 score 8 scripts

jwijffels

topicmodels.etm:Topic Modelling in Embedding Spaces

Find topics in texts which are semantically embedded using techniques like word2vec or Glove. This topic modelling technique models each word with a categorical distribution whose natural parameter is the inner product between a word embedding and an embedding of its assigned topic. The techniques are explained in detail in the paper 'Topic Modeling in Embedding Spaces' by Adji B. Dieng, Francisco J. R. Ruiz, David M. Blei (2019), available at <arXiv:1907.04907>.

Maintained by Jan Wijffels. Last updated 3 years ago.

1 stars 2.90 score 32 scripts

rdinnager

fibre:Fast Evolutionary Trait Modelling on Phylogenies using Branch Regression Models

Implements Phylogenetic Branch Regression models which allow for flexible and versatile models of evolution along a phylogeny. The model can be used to detect shifts in rates of evolution along branches. The model uses a continuous and linear model structure and so can be easily combined with other non-phylogenetic statistical structures, as long as they are implemented using the R package INLA. One major uses of this are to condition on phylogeny in a standard regression between two traits, thus 'accounting' for phylogenetic structure in the response variable, similar to how pgls is used but allowing for a more flexible phylogenetic model. This also allows the phylogenetic model to be combined with the spatial models that INLA excels at (and with comparable flexibility to those spatial models).

Maintained by Russell Dinnage. Last updated 4 months ago.

3 stars 2.71 score 34 scripts

mohmedsoudy

ScRNAIMM:Performing Single-Cell RNA-Seq Imputation by Using Mean/Median Imputation

Performing single-cell imputation in a way that preserves the biological variations in the data. The package clusters the input data to do imputation for each cluster, and do a distribution check using the Anderson-Darling normality test to impute dropouts using mean or median (Yazici, B., & Yolacan, S. (2007) <DOI:10.1080/10629360600678310>).

Maintained by Mohamed Soudy. Last updated 1 years ago.

2.70 score

strancsus

scCAN:Single-Cell Clustering using Autoencoder and Network Fusion

A single-cell Clustering method using 'Autoencoder' and Network fusion ('scCAN') Bang Tran (2022) <doi:10.1038/s41598-022-14218-6> for segregating the cells from the high-dimensional 'scRNA-Seq' data. The software automatically determines the optimal number of clusters and then partitions the cells in a way such that the results are robust to noise and dropouts. 'scCAN' is fast and it supports Windows, Linux, and Mac OS.

Maintained by Bang Tran. Last updated 10 months ago.

2.70 score

kalimu

GitAI:Extracts Knowledge from 'Git' Repositories

Scan multiple 'Git' repositories, pull specified files content and process it with large language models. You can summarize the content in specific way, extract information and data, or find answers to your questions about the repositories. The output can be stored in vector database and used for semantic search or as a part of a RAG (Retrieval Augmented Generation) prompt.

Maintained by Kamil Wais. Last updated 1 months ago.

2.70 score 5 scripts

skeydan

torchaudio:R Interface to 'pytorch''s 'torchaudio'

Provides access to datasets, models and processing facilities for deep learning in audio.

Maintained by Sigrid Keydana. Last updated 2 years ago.

2.70 score

frankiethull

ggpal2:"chores" extension ggplot2 LLM assistant

Extension of "chores" FKA "pals", a LLM assistant add-in for RStudio and Positron IDEs built on "ellmer". "ggpal2" is an assistant designed to help with ggplot2.

Maintained by Frank Hull. Last updated 11 days ago.

2 stars 2.60 score 1 scripts

jcheng5

shinychat:Chat UI Component for 'shiny'

Provides a scrolling chat interface with multiline input, suitable for creating chatbot apps based on Large Language Models (LLMs). Designed to work particularly well with the 'elmer' R package for calling LLMs.

Maintained by Joe Cheng. Last updated 3 months ago.

2.33 score 43 scripts

druegamer

deepregression:Fitting Deep Distributional Regression

Allows for the specification of semi-structured deep distributional regression models which are fitted in a neural network as proposed by Ruegamer et al. (2023) <doi:10.18637/jss.v105.i02>. Predictors can be modeled using structured (penalized) linear effects, structured non-linear effects or using an unstructured deep network model.

Maintained by David Ruegamer. Last updated 4 months ago.

2.28 score 63 scripts 1 dependents

dfalbel

madgrad:'MADGRAD' Method for Stochastic Optimization

A Momentumized, Adaptive, Dual Averaged Gradient Method for Stochastic Optimization algorithm. MADGRAD is a 'best-of-both-worlds' optimizer with the generalization performance of stochastic gradient descent and at least as fast convergence as that of Adam, often faster. A drop-in optim_madgrad() implementation is provided based on Defazio et al (2020) <arxiv:2101.11075>.

Maintained by Daniel Falbel. Last updated 4 years ago.

1.70 score 8 scripts

pigian

proteus:Multiform Seq2Seq Model for Time-Feature Analysis

Seq2seq time-feature analysis based on variational model, with a wide range of distributions available for the latent variable.

Maintained by Giancarlo Vercellino. Last updated 5 days ago.

1.52 score 33 scripts

cran

nFunNN:Nonlinear Functional Principal Component Analysis using Neural Networks

Implementation for 'nFunNN' method, which is a novel nonlinear functional principal component analysis method using neural networks. The crucial function of this package is nFunNNmodel().

Maintained by Rou Zhong. Last updated 11 months ago.

1.00 score

cran

engression:Engression Modelling

Fits engression models for nonlinear distributional regression. Predictors and targets can be univariate or multivariate. Functionality includes estimation of conditional mean, estimation of conditional quantiles, or sampling from the fitted distribution. Training is done full-batch on CPU (the python version offers GPU-accelerated stochastic gradient descent). Based on "Engression: Extrapolation for nonlinear regression?" by Xinwei Shen and Nicolai Meinshausen (2023). Also supports classification (experimental). <arxiv:2307.00835>.

Maintained by Nicolai Meinshausen. Last updated 1 years ago.

1.00 score

aholovchak

DistributionIV:Distributional Instrumental Variable (DIV) Model

Distributional instrumental variable (DIV) model for estimation of the interventional distribution of the outcome Y under a do-intervention on the treatment X. Instruments, predictors and targets can be univariate or multivariate. Functionality includes estimation of the (conditional) interventional mean and quantiles, as well as sampling from the fitted (conditional) interventional distribution.

Maintained by Anastasiia Holovchak. Last updated 1 months ago.

1.00 score

pigian

lambdaTS:Variational Seq2Seq Model with Lambda Transformer for Time Series Analysis

Time series analysis based on lambda transformer and variational seq2seq, built on 'Torch'.

Maintained by Giancarlo Vercellino. Last updated 3 years ago.

1.00 score

pigian

spinner:An Implementation of Graph Net Architecture Based on 'torch'

Proposes a 'torch' implementation of Graph Net architecture allowing different options for message passing and feature embedding.

Maintained by Giancarlo Vercellino. Last updated 2 years ago.

1.00 score 4 scripts