Other packages > Find by keyword >

geodl  

Geospatial Semantic Segmentation with Torch and Terra
View on CRAN: Click here


Download and install geodl package within the R console
Install from CRAN:
install.packages("geodl")

Install from Github:
library("remotes")
install_github("cran/geodl")

Install by package version:
library("remotes")
install_version("geodl", "0.2.0")



Attach the package and use:
library("geodl")
Maintained by
Aaron Maxwell
[Scholar Profile | Author Map]
All associated links for this package
First Published: 2024-08-20
Latest Update: 2024-08-20
Description:
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) <doi: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'.
How to cite:
Aaron Maxwell (2024). geodl: Geospatial Semantic Segmentation with Torch and Terra. R package version 0.2.0, https://cran.r-project.org/web/packages/geodl. Accessed 07 Nov. 2024.
Previous versions and publish date:
No previous versions
Other packages that cited geodl R package
View geodl citation profile
Other R packages that geodl depends, imports, suggests or enhances
Complete documentation for geodl
Functions, R codes and Examples using the geodl R package
Full geodl package functions and examples
Downloads during the last 30 days
Get rewarded with contribution points by helping add
Reviews / comments / questions /suggestions ↴↴↴

Today's Hot Picks in Authors and Packages

bacondecomp  
Goodman-Bacon Decomposition
Decomposition for differences-in-differences with variation in treatment timing from Goodman-Bacon ...
Download / Learn more Package Citations See dependency  
nextGenShinyApps  
Craft Exceptional 'R Shiny' Applications and Dashboards with Novel Responsive Tools
Nove responsive tools for designing and developing 'Shiny' dashboards and applications. The scripts ...
Download / Learn more Package Citations See dependency  
robregcc  
Robust Regression with Compositional Covariates
We implement the algorithm estimating the parameters of the robust regression model with composition ...
Download / Learn more Package Citations See dependency  
con2aqi  
Calculate the AQI from Pollutant Concentration
To calculate the AQI (Air Quality Index) from pollutant concentration data. O3, PM2.5, PM10, CO, SO ...
Download / Learn more Package Citations See dependency  

23,092

R Packages

198,677

Dependencies

62,675

Author Associations

23,089

Publication Badges

© Copyright 2022 - present. All right reserved, rpkg.net.  Based in Cambridge, Massachusetts, USA