Re: [R-sig-Geo] regression-kriging and co-kriging (Edzer Pebesma)
On 8/15/19 12:31 PM, Emanuele Barca wrote: > Dear Edzer, > > sorry for bothering you once more but I need to be sure about my R script. > > In summary, i'm comparing the performance of regression-kriging and > collocated co-kriging. > > Regression-kriging was based on an unique covariate, the elevation Z. > > I use Z as unique ancillary variable in the Co-kriging. > > As first attempt, the final raster maps were completely different. It > appeared that it was due > > to the fact that the dataset was non-stationary and only > regression-kriging overcomes this issue, while co-kriging not. > > But if I pass to universal co-kriging introducing Z as covariate, it > bacomes useless as ancillary variable! > > What is my mistake? You assume that someone else can be sure about your analysis by looking at your R script without having access to your data. And you are starting a personal dialogue on a list, essentially uninviting everyone else to get involved. > > emanuele > > > > > Il 2019-08-15 12:00 r-sig-geo-requ...@r-project.org ha scritto: >> Send R-sig-Geo mailing list submissions to >> r-sig-geo@r-project.org >> >> To subscribe or unsubscribe via the World Wide Web, visit >> https://stat.ethz.ch/mailman/listinfo/r-sig-geo >> or, via email, send a message with subject or body 'help' to >> r-sig-geo-requ...@r-project.org >> >> You can reach the person managing the list at >> r-sig-geo-ow...@r-project.org >> >> When replying, please edit your Subject line so it is more specific >> than "Re: Contents of R-sig-Geo digest..." >> >> >> Today's Topics: >> >> 1. Error running codes (Enoch Gyamfi Ampadu) >> 2. Re: regression-kriging and co-kriging (Edzer Pebesma) >> >> -- >> >> Message: 1 >> Date: Thu, 15 Aug 2019 06:51:49 +0200 >> From: Enoch Gyamfi Ampadu >> To: r-sig-geo@r-project.org >> Subject: [R-sig-Geo] Error running codes >> Message-ID: >> >> Content-Type: text/plain; charset="utf-8" >> >> Dear List, >> >> Please I have been running a certain the codes below to read my image, >> shapeffile to enable me classify for cover with Random forest. I have >> gotten to the point of extracting the raster values and the raster >> information. However I keep getting errors. though I have tried different >> combination and other codes like readOGR for importing the training >> polygon. I will be glad if you could be of help as I am new to using R. >> >> #import clip image of study area >> >> TestImg3 <- brick("C:\\Users\\hp\\Desktop\\Data collection\\Nkandla >> Images\\Landsat\\2019\\08052019\\Corrected\\New >> Folder\\Image08052019_Clip.tif") >> >> #Assign name of bands >> names(TestImg3) <- c(paste0("B", 1:7, coll=""), "B9") >> plotRGB(TestImg3, r=4, g=3, b=2, stretch ="lin") >> >> #Import training shapefile >> sample <- shapefile("C:/Users/hp/Desktop/Data collection/Nkandla >> Images/Landsat/2019/08052019/Corrected/Train08052019_2.shp") >> >> responseCol <- "class" >> >> #(I have tried options of changing "class" to "classname" as reflect >> actaul >> name assigned in ArcMap) >> >> # Overlap the sample polygons on the image (combine the class information >> with extracted values). >> >> pixels = data.frame(matrix(vector(), nrow = 0, ncol = >> length(names(img)) + >> 1)) >> for (i in 1:length(unique(sample[[responseCol]]))){ >> category <- unique(sample[[responseCol]])[i] >> categorymap <- sample[sample[[responseCol]] == category,] >> dataSet <- extract(img, categorymap) >> dataSet <- dataSet[!unlist(lapply(dataSet, is.null))] >> if(is(sample, "SpatialPointsDataFrame")){ >> dataSet <- cbind(dataSet, class = as.numeric(category)) >> pixeles <- rbind(pixeles, dataSet) >> } >> if(is(sample, "SpatialPolygonsDataFrame")){ >> dataSet <- lapply(dataSet, function(x){cbind(x, class = >> as.numeric(rep(category, >> >> nrow(x}) >> df <- do.call("rbind", dataSet) >> pixels <- rbind(pixeles, df) >> } >> >> } >> >> THIS IS THE ERROR I GET FROM RUNNING THE ABOVE CODES >> >>> for (i in 1:length(unique(samples[[responseCol]]))){ >> + category <- unique(samples[[responseCol]])[i] >> + categorymap <- samples[samples[[responseCol]] == category,] >> + dataSet <- extract(img, categorymap) >> + >> + if(is(sample, "SpatialPointsDataFrame")){ >> + dataSet <- cbind(dataSet, class = as.numeric(category)) >> + pixeles <- rbind(pixeles, dataSet) >> + } >> + if(is(sample, "SpatialPolygonsDataFrame")){ >> + dataSet <- lapply(dataSet, function(x){cbind(x, class = >> as.numeric(rep(category, >> + >> nrow(x}) >> + df <- do.call("rbind", dataSet) >> + pixels <- rbind(pixeles, df) >> + } >> + >> + } >> Error in y[i, ] : >> cannot get a slot ("Polygons") from an object of type "NULL" >> >> >> Please help me out. >> Thank you. >> >> Best regards, >> >> Enoch >> >> -- >> *Enoch Gyamfi - Ampadu* >> >> *Geography & Environmental Sciences* >> >> *College of Agriculture
Re: [R-sig-Geo] regression-kriging and co-kriging (Edzer Pebesma)
Dear Edzer, sorry for bothering you once more but I need to be sure about my R script. In summary, i'm comparing the performance of regression-kriging and collocated co-kriging. Regression-kriging was based on an unique covariate, the elevation Z. I use Z as unique ancillary variable in the Co-kriging. As first attempt, the final raster maps were completely different. It appeared that it was due to the fact that the dataset was non-stationary and only regression-kriging overcomes this issue, while co-kriging not. But if I pass to universal co-kriging introducing Z as covariate, it bacomes useless as ancillary variable! What is my mistake? emanuele Il 2019-08-15 12:00 r-sig-geo-requ...@r-project.org ha scritto: Send R-sig-Geo mailing list submissions to r-sig-geo@r-project.org To subscribe or unsubscribe via the World Wide Web, visit https://stat.ethz.ch/mailman/listinfo/r-sig-geo or, via email, send a message with subject or body 'help' to r-sig-geo-requ...@r-project.org You can reach the person managing the list at r-sig-geo-ow...@r-project.org When replying, please edit your Subject line so it is more specific than "Re: Contents of R-sig-Geo digest..." Today's Topics: 1. Error running codes (Enoch Gyamfi Ampadu) 2. Re: regression-kriging and co-kriging (Edzer Pebesma) -- Message: 1 Date: Thu, 15 Aug 2019 06:51:49 +0200 From: Enoch Gyamfi Ampadu To: r-sig-geo@r-project.org Subject: [R-sig-Geo] Error running codes Message-ID: Content-Type: text/plain; charset="utf-8" Dear List, Please I have been running a certain the codes below to read my image, shapeffile to enable me classify for cover with Random forest. I have gotten to the point of extracting the raster values and the raster information. However I keep getting errors. though I have tried different combination and other codes like readOGR for importing the training polygon. I will be glad if you could be of help as I am new to using R. #import clip image of study area TestImg3 <- brick("C:\\Users\\hp\\Desktop\\Data collection\\Nkandla Images\\Landsat\\2019\\08052019\\Corrected\\New Folder\\Image08052019_Clip.tif") #Assign name of bands names(TestImg3) <- c(paste0("B", 1:7, coll=""), "B9") plotRGB(TestImg3, r=4, g=3, b=2, stretch ="lin") #Import training shapefile sample <- shapefile("C:/Users/hp/Desktop/Data collection/Nkandla Images/Landsat/2019/08052019/Corrected/Train08052019_2.shp") responseCol <- "class" #(I have tried options of changing "class" to "classname" as reflect actaul name assigned in ArcMap) # Overlap the sample polygons on the image (combine the class information with extracted values). pixels = data.frame(matrix(vector(), nrow = 0, ncol = length(names(img)) + 1)) for (i in 1:length(unique(sample[[responseCol]]))){ category <- unique(sample[[responseCol]])[i] categorymap <- sample[sample[[responseCol]] == category,] dataSet <- extract(img, categorymap) dataSet <- dataSet[!unlist(lapply(dataSet, is.null))] if(is(sample, "SpatialPointsDataFrame")){ dataSet <- cbind(dataSet, class = as.numeric(category)) pixeles <- rbind(pixeles, dataSet) } if(is(sample, "SpatialPolygonsDataFrame")){ dataSet <- lapply(dataSet, function(x){cbind(x, class = as.numeric(rep(category, nrow(x}) df <- do.call("rbind", dataSet) pixels <- rbind(pixeles, df) } } THIS IS THE ERROR I GET FROM RUNNING THE ABOVE CODES for (i in 1:length(unique(samples[[responseCol]]))){ + category <- unique(samples[[responseCol]])[i] + categorymap <- samples[samples[[responseCol]] == category,] + dataSet <- extract(img, categorymap) + + if(is(sample, "SpatialPointsDataFrame")){ + dataSet <- cbind(dataSet, class = as.numeric(category)) + pixeles <- rbind(pixeles, dataSet) + } + if(is(sample, "SpatialPolygonsDataFrame")){ + dataSet <- lapply(dataSet, function(x){cbind(x, class = as.numeric(rep(category, + nrow(x}) + df <- do.call("rbind", dataSet) + pixels <- rbind(pixeles, df) + } + + } Error in y[i, ] : cannot get a slot ("Polygons") from an object of type "NULL" Please help me out. Thank you. Best regards, Enoch -- *Enoch Gyamfi - Ampadu* *Geography & Environmental Sciences* *College of Agriculture, Engineering & Science* *University of KwaZulu-Natal, Westville Campus* *Private Bag X54001* *Durban, South Africa **– 4000**.* *Phone: +27 835 828255* *email: egamp...@gmail.com * *skype: enoch.ampadu* *The highest evidence of nobility is self-control*. *A simple act of kindness creates an endless ripple*. [[alternative HTML version deleted]] -- Message: 2 Date: Thu, 15 Aug 2019 08:33:59 +0200 From: Edzer Pebesma To: r-sig-geo@r-project.org Subject: Re: [R-sig-Geo] regression-kriging and co-kriging Message-ID: Content-Type: text/plain; charset="utf-8" On 8/12/19 8:21 PM, Emanu