Create a report

Source: R/report.R

Create a report composed of elements specified in the parametrization.

genReport(input_source, output, format = "all", endmembers, kmeans = F,
  title = "archeospec", clean_head = NULL, clean_tail = NULL,
  clean_leaps = NULL, file_header = "Wavelength\\t%s",
  wavelength_start = 350, wavelength_end = 2500, intracorrelation = T,
  mutualinfo = T, endmember_names = NULL, endmember_colors = NULL)

Arguments

input_source

The absolute path to a folder which contains .asd files (or subfolders with .asd files). These files must contain the data as a pair wavelength<tab>value.
output

The path for the output folder. It will contain the report files.
format

The report formats to be generated. One option of "html_document", "pdf_document", "word" or "all" (by default). The "pdf_document" will also include the latex sources.
endmembers

It can be a number or a vector of strings. If it is a number, this number of endmembers will be computer from the signatures. If it is a vector of strings, they will be the filenames of the endmembers which will be fixed.
kmeans

Whether to include the clustering plot or not (TRUE / FALSE)
title

The title for the report, included at the begining of them.
clean_head

Optional. If provided, it filters out the wavelengths smaller than clean_head (a number).
clean_tail

Optional. If provided, it filters out the wavelengths greater than clean_tail (a number)
clean_leaps

Optional. If provided, it must be a vector of wavelength values. It will smooth the leaps produced between sensor changes, when using multiple sensors for different wavelength ranges.
file_header

The format of the hehad in .asd files. After the header, the measurements will be in format wavelength<tab>value. By default, the file header is Wavelength\\t%s (note the double backslash to escape the special character bakslash).
wavelength_start

The smallest wavelength of the measured range.
wavelength_end

The greatest wavelength of the measured range.
intracorrelation

Whether to include the intracorrelation plot or not (TRUE / FALSE)
mutualinfo

Whether to include the mutual information plot or not (TRUE / FALSE)
endmember_names

Optional. If provided, it must be a list of names (strings) to be related to selected endmembers, so we can easily recognise the endmembers in the report. If endmembers is not a list of filenames, this parameter is ignored.
endmember_colors

Optional. If provided, it must be a list of colors (strings) to be related to selected endmembers, so we can easily recognise the endmembers in the report. If endmembers is not a list of filenames, this parameter is ignored. To check the available colors, refer to colors.

Value

It creates a report in html, latex, pdf and / or word formats.

Details

First load the data from input source. This must be an ABSOLUTE path to a folder which contains .asd files, or a tree of folders which contains in the end .asd files. These files must contain the data as a pair wavelength<tab>value, and the data must starts after the line file_header (by default "Wavelength<tab>

Then, a filtering process can be applied to the signatures if any of clean_head, clean_tail or clean_leaps are specified. This consists, respectively, on cutting the lower / higher wavelength values. clean_leaps smooth the leaps produced when using multiple sensors for different range of wavelengths, produced between sensor changes.

Finally, the signatures are unmixed. If endmembers is a number, it computes such number of endmembers using the VCA algorithm. If endmembers is a vector of filenames, then those signatures will be selected as endmembers. In this case, we can assign them a representative name and a fixed color (using endmember_names and endmember_colors), so we can easily recognize this signatures in the report. endmember_names is a vector of string names, and endmember_colors is a vector of string colors (check colors).

Different outputs can optionaly be included in the report:

  • kmeans: (TRUE) run kmeans clustering algorithms; (FALSE) use the endmembers as centroids.

  • intracorrelation: included (TRUE) or not (FALSE).

  • mutualinfo: included (TRUE) or not (FALSE).

The report will always include:

  • A brief description of the dataset (list of signatures)

  • The list of parameters used to create the report

  • The signature plot

  • The selected or computed endmembers, in tabular and plot formats

  • The elbow plot

  • The correspondence between endmembers and clusters plot

  • The endmember weights by cluster bar and whisker box plot

  • The residual values for each signature plot

  • The summary table for the signatures and its weights by endmembers

For more details see the doc in [GitHub](https://github.com/jacintoArias/archeospec/).

See also

load_signature_files

Examples

# NOT RUN {
# Path must be absolute
# In windows 10
path <- "C:/Users/(username)/Documents/data/signatures"
# Where signatures contain .asd files and folders with more .asd files

# In linux
path <- "/home/(username)/data/signatures"

path <- "/path/to/asd/folder"

# generate a report computing three endmembers.
genReport(path, endmembers=3)

# generate a report selecting two endmembers.
genReport(path, endmembers=c("almagre.asd.txt", "blanco.asd.txt"))

# generate a report filtering the signatures and computing three endmembers.
genReport(path, clean_head=400, clean_tail=2400, clean_leaps=c(1001, 1831), endmembers=3)

# generate a report filtering the signatures and selecting two endmembers, giving them a name and a color.
endmembers <- c("signature0000.asd", "signature0001.asd")
names <- c("red", "white")
colors <- c("red2", "white")

genReport(path, clean_head=400, clean_tail=2400, clean_leaps=c(1001, 1831),
          endmembers=endmembers, endmember_names=names, endmember_colors=colors)
# }