|
| 1 | +--- |
| 2 | +Title: '.mul()' |
| 3 | +Description: 'Performs element-wise multiplication of two tensors or multiplies a tensor by a scalar.' |
| 4 | +Subjects: |
| 5 | + - 'Computer Science' |
| 6 | + - 'Machine Learning' |
| 7 | +Tags: |
| 8 | + - 'Deep Learning' |
| 9 | + - 'PyTorch' |
| 10 | + - 'Tensors' |
| 11 | +CatalogContent: |
| 12 | + - 'learn-python-3' |
| 13 | + - 'paths/machine-learning' |
| 14 | +--- |
| 15 | + |
| 16 | +The **`.mul()`** function multiplies each element in the input [tensor](https://www.codecademy.com/resources/docs/pytorch/tensors) with another tensor or a scalar and returns a tensor of the same shape as the input (or broadcasted shape when inputs have different shapes). |
| 17 | + |
| 18 | +## Syntax |
| 19 | + |
| 20 | +```pseudo |
| 21 | +torch.mul(input, other, *, out=None) → Tensor |
| 22 | +``` |
| 23 | + |
| 24 | +**Parameters:** |
| 25 | + |
| 26 | +- `input` (Tensor): Input tensor. |
| 27 | +- `other` (Tensor or Integer): The second input can be another tensor or a scalar value. |
| 28 | +- `out` (Tensor, optional): Optional tensor to store the output. |
| 29 | + |
| 30 | +**Return value:** |
| 31 | + |
| 32 | +Returns a new tensor containing the result of the operation, or modifies the `out` tensor if provided. |
| 33 | + |
| 34 | +## Example 1: Applying `.mul()` with a 1D tensor and a scalar |
| 35 | + |
| 36 | +In this example, `.mul()` multiplies a 1D tensor and a scalar: |
| 37 | + |
| 38 | +```py |
| 39 | +import torch |
| 40 | + |
| 41 | +# Create a tensor |
| 42 | +x = torch.tensor([1, 2, 3]) |
| 43 | + |
| 44 | +# Multiply by scalar |
| 45 | +result = torch.mul(x, 2) |
| 46 | + |
| 47 | +print(result) |
| 48 | +``` |
| 49 | + |
| 50 | +The output of this code is: |
| 51 | + |
| 52 | +```shell |
| 53 | +tensor([2, 4, 6]) |
| 54 | +``` |
| 55 | + |
| 56 | +## Example 2: Using `.mul()` with two 1D tensors |
| 57 | + |
| 58 | +In this example, `.mul()` multiplies each element in a 1D tensor with another 1D tensor: |
| 59 | + |
| 60 | +```py |
| 61 | +import torch |
| 62 | + |
| 63 | +# Create a tensor |
| 64 | +x = torch.tensor([1.0, 2.0, 3.0]) |
| 65 | +# Create a second tensor |
| 66 | +y = torch.tensor([4.0, 5.0, 6.0]) |
| 67 | + |
| 68 | +# Multiply by tensor |
| 69 | +result = torch.mul(x, y) |
| 70 | + |
| 71 | +print(result) |
| 72 | +``` |
| 73 | + |
| 74 | +The output of this code is: |
| 75 | + |
| 76 | +```shell |
| 77 | +tensor([ 4., 10., 18.]) |
| 78 | +``` |
| 79 | + |
| 80 | +## Example 3: Using `.mul()` with a 2D tensor and a 1D tensor |
| 81 | + |
| 82 | +In this example, `.mul()` multiplies a 2D tensor with a 1D tensor using broadcasting: |
| 83 | + |
| 84 | +```py |
| 85 | +import torch |
| 86 | + |
| 87 | +# Create a 2x3 tensor |
| 88 | +x = torch.tensor([[1.0, 2.5, 3.0],[4.0, 5.0, 6.5]]) |
| 89 | +# Create a second tensor |
| 90 | +y = torch.tensor([7.0, 8.0, 9.1]) |
| 91 | + |
| 92 | +# Multiply by tensor |
| 93 | +result = torch.mul(x, y) |
| 94 | + |
| 95 | +print(result) |
| 96 | +``` |
| 97 | + |
| 98 | +The output of this code is: |
| 99 | + |
| 100 | +```shell |
| 101 | +tensor([[ 7.0000, 20.0000, 27.3000], |
| 102 | + [28.0000, 40.0000, 59.1500]]) |
| 103 | +``` |
0 commit comments