humanicheck

Helper package to check data being outputted by humaniverse packages meets expectations

Overview

This package exports a single function, humanicheck() which generates a quick and human-readable data quality report directly in your R console.

Installation

You can install the development version of humanicheck from GitHub with:

# install.packages("pak")
pak::pak("humaniverse/humanicheck")

Example Usage

Using the function is straightforward. Simply pass your data frame or tibble to humanicheck().

humanicheck::humanicheck(penguins)

Report Output

The command above will produce the following report in your console:

── Data Quality Report ─────────────────────────────────────────────────────────

── 1. Tibble Check ──

✖ Input is not a tibble. Consider using `tibble::as_tibble()`.

── 2. Object Size Check ──

✔ Object size (17 kB) is within a reasonable limit.

── 3. Duplicated Rows Check ──

✔ No duplicate rows found.

── 4. NA Value Check ──

✖ Found NA values at these locations:
# A tibble: 19 × 2
     row column     
   <int> <chr>      
 1     4 bill_len   
 2   272 bill_len   
 3     4 bill_dep   
 4   272 bill_dep   
 5     4 flipper_len
 6   272 flipper_len
 7     4 body_mass  
 8   272 body_mass  
 9     4 sex        
10     9 sex        
11    10 sex        
12    11 sex        
13    12 sex        
14    48 sex        
15   179 sex        
16   219 sex        
17   257 sex        
18   269 sex        
19   272 sex        

── 5. SF Geometry Check ──

ℹ Skipping SF checks (not an sf object).

── 6. Outlier Check ──

✔ No outliers found.

── End of Report ───────────────────────────────────────────────────────────────

Checks Performed

The humanicheck() function runs the following diagnostic checks:

• Tibble Check: Verifies that the input data is a tibble.
• Object Size: Checks if the in-memory size of the object exceeds 50MB.
• Duplicated Rows: Identifies and displays any fully duplicated rows.
• NA Values: Locates and reports the positions of any NA values.
• SF Geometry: For sf objects, checks if all geometries are of a single, consistent type.
• Outlier Check: Checks and reports names of numeric columns with outliers using the InterQuartile Method.

License

This package is licensed under the MIT License.