Approximate algorithm for allocating bits to a float given the maximum value

Approximate algorithm for allocating bits to a float given the maximum value

allocate_float_bits(nbits, max_value, signed = FALSE)

Arguments

nbits

number of bits for the low-fidelity representation
max_value

the maximum value to be encoded
signed

whether or not signed values will be encoded: default: FALSE

Value

named list of details about the values this float could represent