PyTorch .floor_divide()

In PyTorch, the .floor_divide() function divides the input by other element-wise and rounds each quotient down to the nearest integer, returning the floored result as a tensor. It supports broadcasting, type promotion, and both integer and floating-point operands. The operation can be expressed as:

$$\text{out}_i = \text{floor}(\frac{\text{input}_i}{\text{other}_i})$$

Syntax

torch.floor_divide(input, other, *, out=None) → Tensor

Parameters:

• input: The input tensor(dividend).
• other: Tensor or Number(divisor).
• out (Optional): A tensor to store the output. If provided, the result is written to this tensor.

Return value:

It returns a new tensor of the same shape as the input, containing the result of element-wise division and rounds the result of the division down to the nearest integer.

Example

In this example, we use floor_divide() to perform divison of two tensors x and y :

import torch

x = torch.tensor([4.0, 3.0])

y = torch.tensor([2.0, 2.0])

print(torch.floor_divide(x, y))

Copy to clipboard
Copy to clipboard

The output of this code is:

tensor([2., 1.])

Copy to clipboard
Copy to clipboard