espnet2.asr.transducer.rnnt_multi_blank.utils.rnnt_helper.div_up

source

espnet2.asr.transducer.rnnt_multi_blank.utils.rnnt_helper.div_up(x: int, y: int)

Computes the ceiling division of two integers.

This function takes two integers, x and y, and returns the result of dividing x by y, rounding up to the nearest whole number. This is useful in scenarios where you want to ensure that the result of a division operation does not fall short of the intended number of groups or buckets, especially in applications like batching or memory allocation.

• Parameters:
  • x (int) – The numerator, an integer value that represents the total number to be divided.
  • y (int) – The denominator, a positive integer value by which to divide x. It must not be zero.
• Returns: The ceiling result of the division of x by y.
• Return type: int
• Raises:ZeroDivisionError – If y is zero, as division by zero is undefined.

Examples

>>> div_up(5, 2)
3
>>> div_up(10, 3)
4
>>> div_up(7, 1)
7
>>> div_up(0, 5)
0

NOTE

This function is designed to be used in CUDA kernels and is decorated with @cuda.jit for that purpose. Ensure that this function is called within a proper CUDA context.