rknx / avrbst Goto Github PK

Calculate the leave-n-out cross validation for final GLMM model on limited test sets

1, 5, 10, 20, 50, (100)

plot the values, and once it stabilizes, no need to go higher. Saves number of models to make.

https://stats.stackexchange.com/questions/399557/can-i-use-wilcoxon-signed-rank-paired-test-to-compare-binary-data

The linters can be specified by creating the .lintr file at the project or home directory. Details can be found at lintr documentation.

Or better

source("c:\data\rprojects\functions\prereqs.R) to rprofile

Use commented lines of - and = to break up your file into easily readable chunks.

#--------------------------------------------

The TSS are much higher this way

Use 10x10 area for each tree
Use geom polygon, rect or ribbon

Cumulative disease Index

Temp:
< 20 : 0
20-25: 1
25-32: 2
32-34: 1
34+: 0

Rain:
0: 0
0.0.1 - 0.1: 1
1+ : 2

Humidity:
0-60: 0
610-70: 0.5
70-80: 1
80:90: 1.5
90-100: 2

Wind:
0.5: 0
5-7: 1
7+: 0

For cross validation phase:

• Transfer dataset to remote location (Hipergator)
• install packages parallel, foreach and rslurm
• in a script
  • make functions to execute
  • make parameter list (one per slurm submission)
  • set slurm options
  • execute
• copy rda back

Add options(help_type = "html")
to ~/.Rprofile to make it persistent

filter <- function(.data, ...) {
cond <- deparse_dots(...)
paste(, collapse = " & ")
.data[with(
.data, eval(
parse(
text = conditions
)
)
), ]
}

1. Fit presence ~ week + dis + gene + gene:dis + gene:week + dir for field data.
2. Get residuals.
3. Merge residuals to each env dataframe and calculate fit.
4. Determine best lag period.

make .bash_login
Export display

Make vcxsrv multiwindow

Install x11

https://youtu.be/6_mbd1hvUnE