Understanding R’s Error in min(c(bnd$x, bnd$y), na.rm = TRUE)

Introduction

The given error occurs when using the min function with a binary operator (c) and na.rm = TRUE. In this blog post, we’ll explore the root of this issue and provide solutions to resolve it.

The Issue

ctd_mba_bound <- ctd_mba[inSide(bounding_box_list, v, w),]

The error occurs when trying to find the minimum value between two vectors x and y. However, in the provided code snippet, both v and w are numeric values. We can use inSide function to check which pixels are inside the bounding box, but the issue lies elsewhere.

What Does This Mean?

In R, the binary operator is used for logical operations (AND (&) and OR (|)), not for mathematical operations like finding minimum or maximum values.

The error message indicates that there’s a non-numeric argument to the binary operator. In this case, it’s likely due to one of the vectors v or w being non-numeric.

Possible Causes

1. Non-Numeric Values: Although the question doesn’t explicitly state where the issue lies, it can be inferred from the provided code snippet that there might be a problem with one of the numeric variables used in the bounding box calculations.
2. Non-Integer Values: The bounding box calculations are likely creating non-integer values for v and w, which could lead to an error when trying to use them as indices.

Solution

Check for Non-Numeric Values

We can check if there are any non-numeric values in the vectors v and w.

# Check for non-numeric values
ctd_mba_bound <- ctd_mba[inSide(bounding_box_list, v, w),]
if (any(is.na(ctd_mba_bound[v])) | any(is.na(ctd_mba_bound[w]))) {
    print("Non-numeric value found")
}

If non-numeric values are present, we need to identify and remove them.

Remove Non-Numeric Values

To do this, you can use the is.numeric function.

# Check for non-numeric values
v_numeric <- v[!is.na(v)]
w_numeric <- w[!is.na(w)]

ctd_mba_bound <- ctd_mba[inSide(bounding_box_list, v_numeric, w_numeric),]

Check for Non-Integer Values

Another possible cause of the issue could be non-integer values in v and w.

# Check for non-integer values
ctd_mba_bound <- ctd_mba[inSide(bounding_box_list, v %in% integer(v), w %in% integer(w)),]

Solution

If the issue is caused by non-numeric or non-integer values in v and w, you can use the code snippet above to remove them.

However, if there’s still an error after removing these values, we need to check the data type of the variables involved in the bounding box calculations.

For example, let’s consider how bounding_box_list is created:

# Create bounding box list
bounding_box_list <- list(
    bounding_box = list(v = ctd_mba$latitude,
                        w = ctd_mba$depth),
    names(bounding_box) = c("v", "w")
)

In this case, the issue could be that ctd_mba$latitude and ctd_mba$depth are not numeric.

# Check data type of variables involved in bounding box calculations
str(ctd_mba$latitude)
str(ctd_mba$depth)

if (!is.numeric(ctd_mba$latitude) | !is.numeric(ctd_mba$depth)) {
    print("Error: Latitude or depth is not numeric")
}

Solution

If the data type of ctd_mba$latitude and ctd_mba$depth is not numeric, you can convert them to integers using the following code:

# Convert latitude and depth to integers
ctd_mba$latitude <- as.integer(ctd_mba$latitude)
ctd_mba$depth <- as.integer(ctd_mba$depth)

ctd_mba_bound <- ctd_mba[inSide(bounding_box_list, v = ctd_mba$latitude,
                                  w = ctd_mba$depth),]

By following these steps and using the provided code snippets, you should be able to identify and resolve the issue that’s causing the error in min(c(bnd$x, bnd$y), na.rm = TRUE).

Last modified on 2023-06-11