Week 4 materials ready to go

week4/index.qmd

@@ -22,8 +22,8 @@ schedule |>
 * Debugging
 * Measuring performance
 * Efficient R pgoramming
-* R environments
 * Caching
+* R environments
 
 ## Online resources
 

week4/slides.qmd

@@ -60,6 +60,7 @@ The **reprex** package helps create *minimal reproducible examples*.
 
 
 ## Debugging tools in R
+\vspace*{-0.2cm}
 \fontsize{14}{15}\sf
 
 * `traceback`: prints out the function call stack after an error occurs; does nothing if there's no error.
@@ -98,21 +99,21 @@ traceback()
 ```
 
 ## Interactive debugging
+\fontsize{13}{14}\sf
 
 * Using `browser()`
 
-  ```{r, eval = FALSE}
-  g <- function(b) {
+  ```r
+  i <- function(d) {
     browser()
-    h(b)
+    if (!is.numeric(d)) stop("`d` must be numeric", call. = FALSE)
+    d + 10
   }
-  f(10)
   ```
 
 * Setting breakpoints
   * Similar to `browser()` but no change to source code.
-  * Can be set in RStudio by clicking to the left of the line number, or pressing `Shift + F9`.
-
+  * Set in RStudio by clicking to left of line number, or pressing `Shift+F9`.
 * `options(error = browser)`
 
 ## Interactive debugging
@@ -121,6 +122,25 @@ traceback()
 * `undebug()` : removes `browser()` statement.
 * `debugonce()` : same as `debug()`, but removes `browser()` after first run.
 
+## Exercises
+
+1.  What's wrong with this code?\fontsize{10}{10}\sf
+
+    ```{r, error = TRUE}
+    # Multivariate scaling function
+    mvscale <- function(object) {
+      # Remove centers
+      mat <- sweep(object, 2L, colMeans(object))
+      # Scale and rotate
+      S <- var(mat)
+      U <- chol(solve(S))
+      z <- mat %*% t(U)
+      # Return orthogonalized data
+      return(z)
+    }
+    mvscale(mtcars)
+    ```
+
 ## Example
 \vspace*{-0.15cm}
 \centerline{\href{https://posit.co/resources/videos/debugging-techniques-in-rstudio-2/}{\includegraphics[width=16cm, height=20cm]{../screenshots/Amanda_Gadrow.png}}}
@@ -152,27 +172,6 @@ traceback()
 * longer object length is not a multiple of shorter object length
 * package is not available for R version `xx`
 
-## Exercises
-
-1. What's wrong with this code?
-
-\fontsize{10}{10}\sf
-
-```{r, error = TRUE}
-# Multivariate scaling function
-mvscale <- function(object) {
-  # Remove centers
-  mat <- sweep(object, 2L, colMeans(object))
-  # Scale and rotate
-  S <- var(mat)
-  U <- chol(solve(S))
-  z <- mat %*% t(U)
-  # Return orthogonalized data
-  return(z)
-}
-mvscale(mtcars)
-```
-
 ## Non-interactive debugging
 
 * Necessary for debugging code that runs in a non-interactive environment.
@@ -309,11 +308,6 @@ bench::mark(
 
     Repeat using `bench::mark()`. Why are they different?
 
-## Exercises
-
-3. Write a function to find the second largest element of a numeric vector. Test several alternatives, and choose the fastest one.
-
-
 # Improving performance
 
 ## Vectorization
@@ -345,11 +339,7 @@ bench::mark(
 
 ## Exercises
 
-4. How can you use `crossprod()` to compute a weighted sum? How much faster is it than the naive `sum(x * w)`? Why is it faster?
-
-## Exercises
-
-5. Write the following algorithm to estimate $\displaystyle\int_0^1 x^2 dx$ using vectorized code
+3. Write the following algorithm to estimate $\displaystyle\int_0^1 x^2 dx$ using vectorized code
 
 ### Monte Carlo Integration
    a. Initialise: `hits = 0`
@@ -399,6 +389,13 @@ compute()
 compute()
 ```
 
+```{r}
+#| include: false
+#| cache: false
+# Need to explicitly remove results.rds for some reason when doing this in quarto
+file.remove(here::here("cache/results.rds"))
+```
+
 ```{r}
 #| label: cache2
 #| cache: false
@@ -463,7 +460,7 @@ memo_sq(2)
 
 ## Exercises
 
-6. Use `bench::mark()` to compare the speed of `sq()` and `memo_sq()`.
+4. Use `bench::mark()` to compare the speed of `sq()` and `memo_sq()`.
 
 # R environments
 
@@ -494,4 +491,4 @@ memo_sq(2)
 
 ## Exercises
 
-7. Add renv to your Assignment 1 project.
+5. Add renv to your Assignment 1 project. Make sure the packages included are the latest CRAN versions of all packages.

week4/workshop_code.R

@@ -0,0 +1,136 @@
+## Using traceback and debug (Slide 8)
+
+f <- function(a) g(a)
+g <- function(b) h(b)
+h <- function(c) i(c)
+i <- function(d) {
+  if (!is.numeric(d)) stop("`d` must be numeric", call. = FALSE)
+  d + 10
+}
+f("a")
+
+traceback()
+
+options(error = recover)
+f("a")
+
+
+
+# Exercise 1 (slide 11): Multivariate scaling function
+mvscale <- function(object) {
+  # Remove centers
+  mat <- sweep(object, 2L, colMeans(object))
+  # Scale and rotate
+  S <- var(mat)
+  U <- chol(solve(S))
+  z <- mat %*% t(U)
+  # Return orthogonalized data
+  return(z)
+}
+mvscale(mtcars)
+
+
+## Slide 20
+
+library(profvis)
+library(bench)
+f <- function() {
+  pause(0.1)
+  g()
+  h()
+}
+g <- function() {
+  pause(0.1)
+  h()
+}
+h <- function() {
+  pause(0.1)
+}
+
+
+# Slide 21
+
+tmp <- tempfile()
+Rprof(tmp, interval = 0.1)
+f()
+Rprof(NULL)
+writeLines(readLines(tmp))
+
+profvis(f())
+
+## Additional profiling example
+
+Rprof()
+x <- NULL
+for(i in seq(1e5)) {
+  x <- c(x, i)
+}
+Rprof(NULL)
+summaryRprof()
+
+Rprof()
+x <- numeric(1e5)
+for(i in seq(1e5)) {
+  x[i] <- i
+}
+Rprof(NULL)
+summaryRprof()
+
+f <- function(n = 1e5) {
+  x <- NULL
+  for(i in seq(n)) {
+    x <- c(x, i)
+  }
+  return(x)
+}
+
+profvis::profvis(f(1e4))
+
+## Slide 23
+
+x <- rnorm(1e6)
+system.time(min(x))
+system.time(sort(x)[1])
+system.time(x[order(x)[1]])
+
+## Slide 24
+
+bench::mark(
+  min(x),
+  sort(x)[1],
+  x[order(x)[1]]
+)
+
+## Exercise 2: Slide 26
+
+x <- exp(rnorm(1e7))
+system.time(sqrt(x))
+system.time(x^0.5)
+system.time(exp(log(x) / 2))
+
+bench::mark(sqrt(x), x^0.5, exp(log(x) / 2))
+
+## Exercise 3: Slide 27
+
+sort(x, partial = 2)
+
+
+# Caching
+
+compute <- function(...) {
+    xfun::cache_rds(rnorm(6), file = "results.rds", ...)
+}
+compute()
+compute()
+compute(rerun = TRUE)
+compute()
+
+
+library(memoise)
+sq <- function(x) {
+  print("Computing square of 'x'")
+  x**2
+}
+memo_sq <- memoise(sq)
+memo_sq(2)
+memo_sq(2)