_expand_dims
expand_dims
¶
Create a new dimension in tensor
along a given axis
.
This method can be used with Numpy data:
n = np.array([2,7,5])
b = fe.backend.expand_dims(n, axis=0) # [[2, 5, 7]]
b = fe.backend.expand_dims(n, axis=1) # [[2], [5], [7]]
This method can be used with TensorFlow tensors:
t = tf.constant([2,7,5])
b = fe.backend.expand_dims(t, axis=0) # [[2, 5, 7]]
b = fe.backend.expand_dims(t, axis=1) # [[2], [5], [7]]
This method can be used with PyTorch tensors:
p = torch.tensor([2,7,5])
b = fe.backend.expand_dims(p, axis=0) # [[2, 5, 7]]
b = fe.backend.expand_dims(p, axis=1) # [[2], [5], [7]]
Parameters:
Name | Type | Description | Default |
---|---|---|---|
tensor |
Tensor
|
The input to be modified, having n dimensions. |
required |
axis |
int
|
Which axis should the new axis be inserted along. Must be in the range [-n-1, n]. |
1
|
Returns:
Type | Description |
---|---|
Tensor
|
A concatenated representation of the |
Raises:
Type | Description |
---|---|
ValueError
|
If |