Calculate the expected number of frames in a time bin.

A convenience function for checking whether a binned results data.table has the right number of frames. Frames per bin = (binwidth_min * 60) / framelength_s. Compare against the frames column of a binned result to identify bins with missing audio (e.g. recorder dropouts or files shorter than the bin width).

Usage

frames_expected(binwidth_min, framelength_s)

Arguments

binwidth_min

Numeric. Bin width in minutes.

framelength_s

Numeric. Frame length in seconds. For model_general_v3 this is 0.96.

Value

A single numeric value: the expected number of frames per full bin.

Examples

# model_general_v3 uses 0.96-second frames
frames_expected(binwidth_min = 5,  framelength_s = 0.96)  # 312.5
#> [1] 312.5
frames_expected(binwidth_min = 20, framelength_s = 0.96)  # 1250
#> [1] 1250
frames_expected(binwidth_min = 60, framelength_s = 0.96)  # 3750
#> [1] 3750

# Check for bins with incomplete audio coverage
if (FALSE) { # \dontrun{
binned <- bin_directory(dir, thresholds = c(ins_buzz = -1.2), binwidth = 20)
expected <- frames_expected(20, 0.96)
binned[frames < expected * 0.9]  # bins with > 10% missing audio
} # }