--- /dev/null
+Package: epclust
+Title: Clustering individual electricity power curves
+Description: EPCLUST: Electric Power curves CLUSTering, through their wavelets decomposition.
+ The main method 'epclust' takes (usually long) time-series in input, and return as many
+ clusters centers as requested, along with their identifiers (if aplicable).
+ Several parameters can be tuned: please refer to the package vignette.
+Version: 0.1-0
+Author:
+ Benjamin Auder <Benjamin.Auder@math.u-psud.fr> [aut,cre],
+ Jairo Cugliari <Jairo.Cugliari@univ-lyon2.fr> [aut]
+Maintainer: Benjamin Auder <Benjamin.Auder@math.u-psud.fr>
+Depends:
+ R (>= 3.0.0)
+Imports:
+ MASS
+Suggests:
+ testthat,
+ parallel,
+ knitr
+License: MIT + file LICENSE
+VignetteBuilder: knitr
--- /dev/null
+YEAR: 2016-2017
+COPYRIGHT HOLDER: Jairo CUGLIARI
--- /dev/null
+epclust = function(data=NULL, con=NULL, raw=FALSE, K, nbPerChunk, ...)
+{
+
+
+#TODO: just a wrapper which calls ppam.exe (system("...")) and reads output (binary) file to retrieve medoids + IDs
+ #on input: can be data or con; data handled by writing it to file (ascii or bin ?!),
+ #con handled
+
+
+ #options for tmp files: in RAM, on disk, on DB (can be distributed)
+
+
+
+ if (!is.null(data))
+ {
+ #full data matrix
+
+ } else if (!is.null(con))
+ {
+ #incremental connection
+ #read it one by one and get coeffs until nbSeriesPerChunk
+ #then launch a clustering task............
+ } else
+ stop("at least 'data' or 'con' argument must be present")
+
+}
--- /dev/null
+# Clustering individual power curves using wavelets decomposition
+
+This package was funded by a contract with EDF; its goal is to identify groups of customers.
+
+## Description
+
+TODO
+
+## Example
+
+TODO
--- /dev/null
+\name{epclust-package}
+\alias{epclust-package}
+\alias{epclust}
+\docType{package}
+
+\title{
+ \packageTitle{epclust}
+}
+
+\description{
+ \packageDescription{epclust}
+}
+
+\details{
+ The package devtools should be useful in development stage, since we rely on testthat for
+ unit tests, and roxygen2 for documentation. knitr is used to generate the package vignette.
+ Concerning the other suggested packages, 'parallel' can speed up (...TODO...)
+
+ The main function is located in R/main.R: it runs the clustering task (TODO: explain more).
+}
+
+\author{
+ \packageAuthor{epclust}
+
+ Maintainer: \packageMaintainer{epclust}
+}
+
+%\references{
+% TODO: Literature or other references for background information
+%}
+
+%\examples{
+% TODO: simple examples of the most important functions
+%}
--- /dev/null
+library(testthat)
+library(epclust)
+
+test_check("epclust")