isaakiel / quantaar Goto Github PK

Should we set up a lintr-bot for automatic syntax-correction?

Never personally worked with it, but I've seen it some times. Facilitates keeping code clean.

Which methods of varnastat should be included, and in which way this should be done?

Options:

• dependency on the package
• Copy code
• Reprogramming of methods

I have queried my swiss contacts, who started using R recently, which features they would love to have in an archaeological package, and here is what they replied (I just put the answer here like I received it in plain german for discussion):

===snip===

• check eines reingeladenen csv auf doppelte rownames
• scatterplotmatix, welche die strukturierenden Elemente (bei XRF) automatisch markiert (jetzt drucke ich das aus und markiere mir das von Hand auf Papier)
• nach der automatischen Erkennung der strukturierenden Variablen könnten diese gleich ausgewählt und in eine PCA gefüttert werden: am besten gleich umskaliert, skaliert und logarithmiert
• dann markieren der Ausreisser und in einem nächsten Schritt diese weglassen (bzw. PCA ohne Aureisser wiederholen)
• dann gleiches mit Clusteranalysen, dort dann die Gruppen, beim gewünschten Knie isolieren, eine Spalte in die Datentabelle einfügen bei der dann Gruppennamen für die Clusters vergeben werden
• Daten nach Gruppennamen sortieren und gleich in LDA füttern

(die letzten beiden schritte mache ich zurzeit noch in Excel...)

• wichtig bei allen Plots: immer gleich die Legenden nett platzieren können, Farben für Gruppen selber wählen können, Individuennummern anschreiben
• generell auch immer gleich Werte anzeigen lassen, z. B. des Medians bei einem Boxplot etc.
• automatischer Export von Plots in einen bestimmten Ordner, wobei der Pfad zu R-script (inklusive Zeile darin) und das Erstelldatum auf dem Plot angegeben werden (zwecks Übersichtlichkeit)

===snip===

I think it's time to decide what we want to do with this package. When we started this 2 years ago we had a lot of ideas, but no concrete plan and therefore quantAAR remained a lazy dumping ground for old code. Some of the ideas we had back then became reality in the meantime (e.g. ISAAKiel/oxcAAR, nfrerebeau/tabula) and others remained unsolved.

I suggest to focus with this package on tidy wrappers and utility functions for the most important multivariate analysis methods in archaeology. I did some work towards this idea and want to hear your opinion. What I want to do:

1. Closing all current issues (#2, #3, #4, #7)
2. Cleaning and better documentation + tests + README of the currently available utility functions for correspondence analysis and seriation.
3. CRAN release with this small set of features in the following weeks.

I'm especially interested in your thoughts @MartinHinz .

Should 14C calibration be included at all?

If so, should we depend on eg. bchron, or shall we code a calibration by ourselves?

If so, including Bayesian Calibration? Maybe as future milestone?

I'm currently working on some tools to work with 3D spatial data within the scope of excavations. This could be part of a bigger effort to provide tools for field archaeologists. Some ideas:

• interfaces for measurement gear (COM-Port)
• analysis and visualization of 3D point distributions
  • measurements
  • SFM
  • etc.
• raster analysis
  • SFM pictures quality assesement
  • semi automatic classification for profile pictures
  • etc.

Include c(c)a and/or other ordination methods like (N)MDS.

I opt for dependency on vegan.

The aim should be to make them more accessible for archaeologists...

I suggest implementing a set of functions for a standard workflow of burial site analysis.
This could involve

• a data structure describing a cemetery (I know, you hate lists...)
• summary statistics
• varnastat analysis of the burial items
• measuring inequality using gini and lorenz curve from ineq
• ca and seriation
• clustering
• 'Social Status Analysis' like BASP
• 14C calibration
• ... (add more)

This would include and refer also to feature_requests #2 and #3 ...