The Artificial Neural Networks Algorithm Leabra (leabRa)
The algorithm Leabra (local error driven and associative biologically realistic algorithm) allows for the construction of artificial neural networks that are biologically realistic and balance supervised and unsupervised learning within a single framework. This package is based on the ‘MATLAB’ version by Sergio Verduzco-Flores, which in turn was based on the description of the algorithm by Randall O’Reilly (1996) <>. For more general (not ‘R’ specific) information on the algorithm Leabra see <https://…/Leabra>.

Producing Administrative Boundary Map with Additional Features Embedded (mapReasy)
Produce administrative boundary map, visualize and compare different factors on map, tracking latitude and longitude, bubble plot. The package provides some handy functions to produce different administrative maps easily. Functions to obtain colorful visualization of different regions of interest and sub-divisional administrative map at different levels are included. This csn be used to increase feasibility of mapping disease pattern across different regions (disease mapping) with appropriate colors having intensity coherent with magnitude of prevalence. In many surveys, information on location of sample are collected. Sometimes it is of interest to quick look at the spreadness of the collected sample, check if any observation falls outside of the survey area and identify them. The package provides unique function to perform these tasks easily. Besides, some additional features have been added to make ad-lib comparison of different factors across the region through these maps. Visual presentation of two different variables on a particular map using two way bubble plot is also provided. Simple bar chart and pie chart can be produced on map to compare several factors.This package will be helpful to researchers-both statistician and non-statistician, to create geographic location wise plotting of different indicators. These types of maps are used in different research areas such public health, economics, environment, journalism etc. It provides functions that will also be helpful to users to create map using two indicators at a time (for example, shade on a map will give the information of one indicator variable, bar/pie/bubble chart will give the information on another indicator). Users only need to select the indicator’s value and country wise region specific shapefile and run the functions to find their graphs quickly.The distinguishable features of the functions in this package are they are easy to understand to new R users who are searching some ad-lib functions to produce administrative map with different features and easy to use for those who are unfamiliar with file format of spatial data or geographic location data. Functions in this package adopt, compile and implement functions from some well-known packages on handling spatial data to make an user friendly functionality. So users do not need any additional knowledge about spatial statistics or geographic location data. All the examples presented in this package use shapefile of country Bangladesh downloaded from <>. Users are requested to visit <>, then select Download, then choose country and shapefile from country and File format dropdown menu. After downloading the shapefile of any particular country as compressed file, unzip the file and keep them in a known directory or working directory. Shapefiles of respective countries will be required to produce corresponding country maps. Use shapefile of corresponding country to produce all types of maps available in this package.

Support for the ‘Mathpix’ API (Image to ‘LaTeX’) (mathpix)
Given an image of a formula (typeset or handwritten) this package provides calls to the ‘Mathpix’ service to produce the ‘LaTeX’ code which should generate that image, and pastes it into a (e.g. an ‘rmarkdown’) document. See <https://…/> for full details. ‘Mathpix’ is an external service and use of the API is subject to their terms and conditions.

Convert Html into Text (htm2txt)
Wipe out tags in a html document and extract a text from it.

Accelerated Sparse Discriminant Analysis (accSDA)
Implementation of sparse linear discriminant analysis, which is a supervised classification method for multiple classes. Various novel optimization approaches to this problem are implemented including alternating direction method of multipliers (ADMM), proximal gradient (PG) and accelerated proximal gradient (APG) (See Atkins et al. <arXiv:1705.07194>). Functions for performing cross validation are also supplied along with basic prediction and plotting functions. Sparse zero variance discriminant analysis (SZVD) is also included in the package (See Ames and Hong, <arXiv:1401.5492>). See the github wiki for a more extended description.

Functional Regression using Signal Compression Approach (FRegSigCom)
Signal compression methods for function-on-function (FOF) regression with functional response and functional predictors, including linear models with both scalar and functional predictors for a small number of functional predictors, linear models with functional predictors for a large number of functional predictors, and nonlinear models. Ruiyan Luo and Xin Qi (2017) <doi:10.1080/01621459.2016.1164053>.