Erik-Jan van Kesteren vankesteren

Department compute server

This is the documentation for using the Methods & Statistics department compute server.

	rtrunc <- function(rdist, min, max) {
	# deal with pipe and also with normal evaluation
	# https://github.com/tidyverse/magrittr/issues/115#issuecomment-173894787
	parents <- lapply(sys.frames(), parent.env)
	is_magrittr_env <- vapply(parents, identical, logical(1), y = environment(`%>%`))
	if (any(is_magrittr_env)) {
	distcall <- get("lhs", sys.frames()[[base::max(which(is_magrittr_env))]])
	} else {
	distcall <- substitute(rdist)
	}

	library(tidyverse)
	library(sf)
	library(osmdata)
	library(osmenrich)
	library(patchwork)

	# conditional IDW functions

	conditional_idw <- function(formula, data, p = 1, max.iter = 100, tol = 1e-10) {

	library(sf)
	library(httr)
	library(tidyverse)


	url <- parse_url("https://geodata.nationaalgeoregister.nl/bag/wfs/v1_1")

	postcode_filter <- function(postcode, gebruiksdoel = "woonfunctie") {
	sprintf(
	"<Filter>

	library(deldir)
	library(ggplot2)
	library(firatheme)

	N <- 40
	dat <- matrix(rnorm(N*2), N)
	dat <- dat %% chol(matrix(c(1, 0, 0, 1.5), 2)) %% chol(solve(cov(dat)))
	dat2 <- deldir(dat[,1], dat[,2])
	dat1 <- data.frame(dat)
	colnames(dat1) <- c("x1", "y1")

	# script that outputs a graph
	library(tidyverse)
	library(firatheme)
	wine <- read_delim("https://archive.ics.uci.edu/ml/machine-learning-databases/wine/wine.data",
	delim = ",",
	col_names = c(
	"Cultivar", "Alcohol", "Malic acid", "Ash", "Alcalinity of ash", "Magnesium",
	"Total phenols", "Flavanoids", "Nonflavanoid phenols", "Proanthocyanins",
	"Color intensity", "Hue", "OD280/OD315", "Proline"
	)

	module Adamopt

	# This is a module implementing vanilla Adam (https://arxiv.org/abs/1412.6980).
	export Adam, step!

	# Struct containing all necessary info
	mutable struct Adam
	theta::AbstractArray{Float64} # Parameter array
	loss::Function # Loss function
	grad::Function # Gradient function

	# Sparse coding for MNIST feature extraction using autodiff
	using Zygote: gradient
	using MLDatasets: MNIST
	using LinearAlgebra: Diagonal
	using ImageCore

	# Goodfellow page 629, equation 19.16, but per pixel
	function loss(H::Matrix{Float64}, W::Matrix{Float64})
	(sum(abs.(H)) + sum((X - H*W).^2)) / L
	end