Merge pull request #574 from UBC-DSCI/bug-hunt

Bug Hunt

source/classification1.Rmd

@@ -324,7 +324,7 @@ Figure \@ref(fig:05-knn-1).
 perim_concav_with_new_point <-  bind_rows(cancer,
                                           tibble(Perimeter = new_point[1],
                                                  Concavity = new_point[2],
-                                                 Class = "unknown")) |>
+                                                 Class = "Unknown")) |>
   ggplot(aes(x = Perimeter,
              y = Concavity,
              color = Class,
@@ -379,7 +379,7 @@ not, if you consider the other nearby points.
 perim_concav_with_new_point2 <- bind_rows(cancer,
                                           tibble(Perimeter = new_point[1],
                                                  Concavity = new_point[2],
-                                                 Class = "unknown")) |>
+                                                 Class = "Unknown")) |>
   ggplot(aes(x = Perimeter,
              y = Concavity,
              color = Class,
@@ -466,7 +466,7 @@ In order to find the $K=5$ nearest neighbors, we will use the `slice_min` functi
 perim_concav <- bind_rows(cancer,
                           tibble(Perimeter = new_point[1],
                                  Concavity = new_point[2],
-                                 Class = "unknown")) |>
+                                 Class = "Unknown")) |>
   ggplot(aes(x = Perimeter,
              y = Concavity,
              color = Class,
@@ -945,12 +945,12 @@ Standardizing your data should be a part of the preprocessing you do
 before predictive modeling and you should always think carefully about your problem domain and
 whether you need to standardize your data.
 
-```{r 05-scaling-plt, echo = FALSE, fig.height = 4, fig.align = "center", fig.cap = "Comparison of K = 3 nearest neighbors with standardized and unstandardized data."}
+```{r 05-scaling-plt, echo = FALSE, fig.height = 4, fig.align = "center", fig.cap = "Comparison of K = 3 nearest neighbors with unstandardized and standardized data."}
 
 attrs <- c("Area", "Smoothness")
 
 # create a new obs and get its NNs
-new_obs <- tibble(Area = 400, Smoothness = 0.135, Class = "unknown")
+new_obs <- tibble(Area = 400, Smoothness = 0.135, Class = "Unknown")
 my_distances <- table_with_distances(unscaled_cancer[, attrs],
                                      new_obs[, attrs])
 neighbors <- unscaled_cancer[order(my_distances$Distance), ]
@@ -989,7 +989,7 @@ unscaled <- ggplot(unscaled_cancer, aes(x = Area,
   ), color = "black", linewidth = 0.5, show.legend = FALSE)
 
 # create new scaled obs and get NNs
-new_obs_scaled <- tibble(Area = -0.72, Smoothness = 2.8, Class = "unknown")
+new_obs_scaled <- tibble(Area = -0.72, Smoothness = 2.8, Class = "Unknown")
 my_distances_scaled <- table_with_distances(scaled_cancer[, attrs],
                                             new_obs_scaled[, attrs])
 neighbors_scaled <- scaled_cancer[order(my_distances_scaled$Distance), ]
@@ -1067,7 +1067,7 @@ ggplot(unscaled_cancer, aes(x = Area,
    facet_zoom(x = ( Area > 380 & Area < 420) ,
               y = (Smoothness > 0.08 & Smoothness < 0.14), zoom.size = 2) +
     theme_bw() +
-    theme(text = element_text(size = 18), axis.title=element_text(size=18), legend.position="bottom")
+    theme(text = element_text(size = 13), axis.title=element_text(size=13), legend.position="bottom")
 ```
 
 ### Balancing
@@ -1141,7 +1141,7 @@ neighbors <- rare_cancer[order(my_distances$Distance), ]
 rare_plot <- bind_rows(rare_cancer,
                        tibble(Perimeter = new_point[1],
                               Concavity = new_point[2],
-                              Class = "unknown")) |>
+                              Class = "Unknown")) |>
   ggplot(aes(x = Perimeter, y = Concavity, color = Class, shape = Class)) +
   geom_point(alpha = 0.5) +
   labs(color = "Diagnosis",
@@ -1175,8 +1175,8 @@ rare_plot + geom_point(aes(x = new_point[1], y = new_point[2]),
 ```
 
 Figure \@ref(fig:05-upsample-2) shows what happens if we set the background color of
-each area of the plot to the predictions the K-nearest neighbors
-classifier would make. We can see that the decision is
+each area of the plot to the prediction the K-nearest neighbors
+classifier would make for a new observation at that location. We can see that the decision is
 always "benign," corresponding to the blue color.
 
 ```{r 05-upsample-2, echo = FALSE, fig.height = 3.5, fig.width = 4.5, fig.align = "center", fig.cap = "Imbalanced data with background color indicating the decision of the classifier and the points represent the labeled data."}

source/classification2.Rmd

@@ -139,7 +139,7 @@ it classified 3 malignant observations as benign, and 4 benign observations as
 malignant. The accuracy of this classifier is roughly
 89%, given by the formula
 
-$$\mathrm{accuracy} = \frac{\mathrm{number \; of  \; correct  \; predictions}}{\mathrm{total \;  number \;  of  \; predictions}} = \frac{1+57}{1+57+4+3} = 0.892$$
+$$\mathrm{accuracy} = \frac{\mathrm{number \; of  \; correct  \; predictions}}{\mathrm{total \;  number \;  of  \; predictions}} = \frac{1+57}{1+57+4+3} = 0.892.$$
 
 But we can also see that the classifier only identified 1 out of 4 total malignant
 tumors; in other words, it misclassified 75% of the malignant cases present in the
@@ -245,7 +245,7 @@ Here, we pass in the number `1`.
 
 ```{r}
 set.seed(1)
-random_numbers1 <- sample(0:9, 10, replace=TRUE)
+random_numbers1 <- sample(0:9, 10, replace = TRUE)
 random_numbers1
 ```
 
@@ -255,7 +255,7 @@ we run the `sample` function again, we will
 get a fresh batch of 10 numbers that also look random.
 
 ```{r}
-random_numbers2 <- sample(0:9, 10, replace=TRUE)
+random_numbers2 <- sample(0:9, 10, replace = TRUE)
 random_numbers2
 ```
 
@@ -265,10 +265,10 @@ value.
 
 ```{r}
 set.seed(1)
-random_numbers1_again <- sample(0:9, 10, replace=TRUE)
+random_numbers1_again <- sample(0:9, 10, replace = TRUE)
 random_numbers1_again
 
-random_numbers2_again <- sample(0:9, 10, replace=TRUE)
+random_numbers2_again <- sample(0:9, 10, replace = TRUE)
 random_numbers2_again
 ```
 
@@ -278,19 +278,19 @@ obtain a different sequence of random numbers.
 
 ```{r}
 set.seed(4235)
-random_numbers <- sample(0:9, 10, replace=TRUE)
-random_numbers
+random_numbers1_different <- sample(0:9, 10, replace = TRUE)
+random_numbers1_different
 
-random_numbers <- sample(0:9, 10, replace=TRUE)
-random_numbers
+random_numbers2_different <- sample(0:9, 10, replace = TRUE)
+random_numbers2_different
 ```
 
 In other words, even though the sequences of numbers that R is generating *look*
 random, they are totally determined when we set a seed value!
 
 So what does this mean for data analysis? Well, `sample` is certainly
 not the only function that uses randomness in R. Many of the functions
-that we use in `tidymodels`, `tidyverse`, and beyond use randomness&mdash;many of them
+that we use in `tidymodels`, `tidyverse`, and beyond use randomness&mdash;some of them
 without even telling you about it. So at the beginning of every data analysis you
 do, right after loading packages, you should call the `set.seed` function and
 pass it an integer that you pick.
@@ -512,10 +512,10 @@ cancer_acc_1 <- cancer_test_predictions |>
                 filter(.metric == 'accuracy')
 
 cancer_prec_1 <- cancer_test_predictions |>
-  precision(truth = Class, estimate = .pred_class, event_level="first")
+  precision(truth = Class, estimate = .pred_class, event_level = "first")
 
 cancer_rec_1 <- cancer_test_predictions |>
-  recall(truth = Class, estimate = .pred_class, event_level="first")
+  recall(truth = Class, estimate = .pred_class, event_level = "first")
 ```
 
 In the metrics data frame, we filtered the `.metric` column since we are
@@ -537,12 +537,12 @@ If the labels were in the other order, we would instead use `event_level="second
 
 ```{r 06-precision}
 cancer_test_predictions |>
-  precision(truth = Class, estimate = .pred_class, event_level="first")
+  precision(truth = Class, estimate = .pred_class, event_level = "first")
 ```
 
 ```{r 06-recall}
 cancer_test_predictions |>
-  recall(truth = Class, estimate = .pred_class, event_level="first")
+  recall(truth = Class, estimate = .pred_class, event_level = "first")
 ```
 
 The output shows that the estimated precision and recall of the classifier on the test data was
@@ -1400,6 +1400,7 @@ res <- tibble(ks = ks, accs = accs, fixedaccs = fixedaccs, nghbrs = nghbrs)
 
 plt_irrelevant_accuracies <- ggplot(res) +
               geom_line(mapping = aes(x=ks, y=accs)) +
+  	      geom_point(mapping = aes(x=ks, y=accs)) +
               labs(x = "Number of Irrelevant Predictors",
                    y = "Model Accuracy Estimate") +
   theme(text = element_text(size = 18), axis.title=element_text(size=18))
@@ -1420,9 +1421,10 @@ this evidence; if we fix the number of neighbors to $K=3$, the accuracy falls of
 
 ```{r 06-neighbors-irrelevant-features, echo = FALSE, warning = FALSE, fig.retina = 2, out.width = "65%", fig.align = "center", fig.cap = "Tuned number of neighbors for varying number of irrelevant predictors."}
 plt_irrelevant_nghbrs <- ggplot(res) +
+              geom_point(mapping = aes(x=ks, y=nghbrs)) +
               geom_line(mapping = aes(x=ks, y=nghbrs)) +
               labs(x = "Number of Irrelevant Predictors",
-                   y = "Number of neighbors") +
+                   y = "Tuned number of neighbors") +
   theme(text = element_text(size = 18), axis.title=element_text(size=18))
 
 plt_irrelevant_nghbrs
@@ -1434,6 +1436,7 @@ res_tmp <- res %>% pivot_longer(cols=c("accs", "fixedaccs"),
                                 values_to="accuracy")
 
 plt_irrelevant_nghbrs <- ggplot(res_tmp) +
+              geom_point(mapping = aes(x=ks, y=accuracy, color=Type)) +
               geom_line(mapping = aes(x=ks, y=accuracy, color=Type)) +
               labs(x = "Number of Irrelevant Predictors", y = "Accuracy") +
               scale_color_manual(labels= c("Tuned K", "K = 3"), values = c("darkorange", "steelblue")) +
@@ -1661,6 +1664,7 @@ where the elbow occurs, and whether adding a variable provides a meaningful incr
 
 fwd_sel_accuracies_plot <- accuracies |>
   ggplot(aes(x = size, y = accuracy)) +
+  geom_point() +
   geom_line() +
   labs(x = "Number of Predictors", y = "Estimated Accuracy")  +
   theme(text = element_text(size = 20), axis.title=element_text(size=20))

source/inference.Rmd

@@ -83,7 +83,7 @@ In general, the process of using a sample to make a conclusion about the
 broader population from which it is taken is referred to as **statistical inference**.
 \index{inference}\index{statistical inference|see{inference}}
 
-```{r 11-population-vs-sample, echo = FALSE, message = FALSE, warning = FALSE, fig.align = "center", fig.cap = "Population versus sample.", out.width="100%"}
+```{r 11-population-vs-sample, echo = FALSE, message = FALSE, warning = FALSE, fig.align = "center", fig.cap = "The process of using a sample from a broader population to obtain a point estimate of a population parameter. In this case, a sample of 10 individuals yielded 6 who own an iPhone, resulting in an estimated population proportion of 60% iPhone owners. The actual population proportion in this example illustration is 53.8%.", out.width="100%"}
 knitr::include_graphics("img/inference/population_vs_sample.png")
 ```
 
@@ -443,7 +443,7 @@ sampling_distribution_40
 In Figure \@ref(fig:11-example-means4), the sampling distribution of the mean
 has one peak and is \index{sampling distribution!shape} bell-shaped. Most of the estimates are between
 about  \$`r round(quantile(sample_estimates$mean_price)[2], -1)` and
-\$`r round(quantile(sample_estimates$mean_price)[4], -1)`; but there are
+\$`r round(quantile(sample_estimates$mean_price)[4], -1)`; but there is
 a good fraction of cases outside this range (i.e., where the point estimate was
 not close to the population parameter). So it does indeed look like we were
 quite lucky when we estimated the population mean with only
@@ -843,8 +843,8 @@ boot20000
 tail(boot20000)
 ```
 
-Let's take a look at histograms of the first six replicates of our bootstrap samples.
-```{r 11-bootstrapping-six-bootstrap-samples, echo = TRUE, fig.pos = "H", out.extra="", message = FALSE, warning = FALSE, fig.align = "center", fig.cap = "Histograms of first six replicates of bootstrap samples."}
+Let's take a look at the histograms of the first six replicates of our bootstrap samples.
+```{r 11-bootstrapping-six-bootstrap-samples, echo = TRUE, fig.pos = "H", out.extra="", message = FALSE, warning = FALSE, fig.align = "center", fig.cap = "Histograms of the first six replicates of the bootstrap samples."}
 six_bootstrap_samples <- boot20000 |>
   filter(replicate <= 6)
 

source/intro.Rmd

@@ -89,7 +89,7 @@ tongues in Canada, and how many people speak each of them?*
 Every good data analysis begins with a *question*—like the
 above—that you aim to answer using data. As it turns out, there
 are actually a number of different *types* of question regarding data:
-descriptive, exploratory, inferential, predictive, causal, and mechanistic,
+descriptive, exploratory, predictive, inferential, causal, and mechanistic,
 all of which are defined in Table \@ref(tab:questions-table).
 Carefully formulating a question as early as possible in your analysis—and
 correctly identifying which type of question it is—will guide your overall approach to
@@ -174,10 +174,12 @@ Since we are using R for data analysis in this book, the first step for us is to
 load the data into R. When we load tabular data into
 R, it is represented as a *data frame* object\index{data frame!overview}. Figure
 \@ref(fig:img-spreadsheet-vs-dataframe) shows that an R data frame is very similar
-to a spreadsheet. We refer to the rows as \index{observation} **observations**; these are the things that we
-collect the data on, e.g., voters, cities, etc. We refer to the columns as \index{variable}
-**variables**; these are the characteristics of those observations, e.g., voters' political
-affiliations, cities' populations, etc.
+to a spreadsheet. We refer to the rows as \index{observation} **observations**;
+these are the individual objects
+for which we collect data. In Figure \@ref(fig:img-spreadsheet-vs-dataframe), the observations are
+languages. We refer to the columns as **variables**; these are the characteristics of each
+observation. In Figure \@ref(fig:img-spreadsheet-vs-dataframe), the variables are the the
+language's category, its name, the number of mother tongue speakers, etc.
 
 ```{r img-spreadsheet-vs-dataframe, echo = FALSE, message = FALSE, warning = FALSE, fig.align = "center", fig.cap = "A spreadsheet versus a data frame in R.", out.width="100%", fig.retina = 2}
 knitr::include_graphics("img/intro/spreadsheet_vs_dataframe.png")
@@ -235,7 +237,7 @@ library(tidyverse)
 ```
 
 > **Note:** You may have noticed that we got some extra
-> output from R saying `Attaching packages` and `Conflicts` below our code
+> output from R regarding attached packages and conflicts below our code
 > line. These are examples of *messages* in R, which give the user more
 > information that might be handy to know. The `Attaching packages` message is
 > natural when loading `tidyverse`, since `tidyverse` actually automatically
@@ -450,7 +452,7 @@ selected_lang <- select(aboriginal_lang, language, mother_tongue)
 selected_lang
 ```
 
-## Using `arrange` to order and `slice` to select rows by index number
+## Using `arrange` to order and `slice` to select rows by index number {#arrangesliceintro}
 
 We have used `filter` and `select` to obtain a table with only the Aboriginal
 languages in the data set and their associated counts. However, we want to know
@@ -498,7 +500,7 @@ counts... But perhaps, seeing these numbers, we became curious about the
 *percentage* of the population of Canada associated with each count. It is
 common to come up with new data analysis questions in the process of answering
 a first one&mdash;so fear not and explore! To answer this small
-question-along-the-way, we need to divide each count in the `mother_tongue`
+question along the way, we need to divide each count in the `mother_tongue`
 column by the total Canadian population according to the 2016
 census&mdash;i.e., 35,151,728&mdash;and multiply it by 100. We can perform
 this computation using the `mutate` function. We pass the `ten_lang`
@@ -521,7 +523,7 @@ as a mother tongue by between 0.008% and 0.18% of the Canadian population.
 
 ## Exploring data with visualizations
 
-We have now answered our initial question by generating the `ten_lang` table!
+The `ten_lang` table we generated in Section \@ref(arrangesliceintro) answers our initial data analysis question.
 Are we done? Well, not quite; tables are almost never the best way to present
 the result of your analysis to your audience. Even the `ten_lang` table with
 only two columns presents some difficulty: for example, you have to scrutinize

source/jupyter.Rmd

@@ -29,15 +29,15 @@ filled_circle <- function(){
   if(is_latex_output()) {
       "\\faCircle{}"
   } else {
-      "\U25EF"
+      "\U2B24"
   }
 }
 
 circle <- function(){
   if(is_latex_output()) {
       "\\faCircle[regular]{}"
   } else {
-      "\U2B24"
+      "\U25EF"
   }
 }
 ```
@@ -68,7 +68,7 @@ By the end of the chapter, readers will be able to do the following:
 
 ## Jupyter
 
-Jupyter is a web-based interactive development environment for creating, editing,
+Jupyter [@kluyver2016jupyter] is a web-based interactive development environment for creating, editing,
 and executing documents called Jupyter notebooks. Jupyter notebooks \index{Jupyter notebook} are
 documents that contain a mix of computer code (and its output) and formattable
 text. Given that they combine these two analysis artifacts in a single

source/reading.Rmd

@@ -352,8 +352,8 @@ Non-Official & Non-Aboriginal languages Arabic  419890  223535  5585    629055
 
 To read this into R using the `read_delim` function, we specify the path
 to the file as the first argument, provide
-the tab character `"\t"` as the `delim` argument \index{read function!delim argument},
-and set the `col_names` argument to `FALSE` to denote that there are no column names
+the tab character `"\t"` as the `delim` argument,
+and set \index{read function!delim argument} the `col_names` argument to `FALSE` to denote that there are no column names
 provided in the data. Note that the `read_csv`, `read_tsv`, and `read_delim` functions
 all have a `col_names` argument \index{read function!col\_names argument} with
 the default value `TRUE`.

source/references.bib

@@ -20,7 +20,7 @@ @article{knncover
   pages = {21--27}}
 
 @article{penguinpaper,
-  title = {Ecological sexual dimorphism and environmental variability within a community of {A}ntarctic penguins (genus \emph{Pygoscelis})},
+  title = {Ecological sexual dimorphism and environmental variability within a community of {A}ntarctic penguins (genus Pygoscelis)},
   year = {2014},
   author = {Kristen Gorman and Tony Williams and William Fraser},
   journal = {PLoS ONE},

source/regression1.Rmd

@@ -107,7 +107,7 @@ is that we are now predicting numerical variables instead of categorical variabl
 
 > **Note:** You can usually tell whether a\index{categorical variable}\index{numerical variable} variable is numerical or
 > categorical—and therefore whether you need to perform regression or
-> classification—by taking two response variables X and Y from your data,
+> classification—by taking the response variable for two observations X and Y from your data,
 > and asking the question, "is response variable X *more* than response
 > variable Y?" If the variable is categorical, the question will make no sense.
 > (Is blue more than red?  Is benign more than malignant?) If the variable is