Conversation
|
rename lp_normalize to normalize, and remove l2_normalize |
|
Should default dim = -1? |
|
@soumith pushed |
|
yes, please put p=2 as the default (you will have to either have a default dim argument, or move p to 3rd argument) |
torch/nn/functional.py
Outdated
| .. math:: | ||
| v = \frac{v}{\max(\lVert v \rVert_p, \epsilon)} | ||
|
|
||
| for each subtensor v over dimension dim of input. |
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
torch/nn/functional.py
Outdated
|
|
||
|
|
||
| def normalize(input, p, dim, eps=1e-12): | ||
| r"""Performs l_p normalization of inputs over specified dimension. |
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
torch/nn/functional.py
Outdated
| return loss | ||
|
|
||
|
|
||
| def normalize(input, p, dim, eps=1e-12): |
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
torch/nn/functional.py
Outdated
| for each subtensor v over dimension dim of input. Each subtensor is flattened into a vector, | ||
| i.e. :math:`\lVert v \rVert_p` is not a matrix norm. | ||
|
|
||
| With default arguments normalizes over the last dimension with Euclidean norm. |
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
* master: Add F.normalize (pytorch#1467) Expose custom attributes from C++ functions (pytorch#1430) Add high order gradient support for Sigmoid (pytorch#1496)
with gradcheck and doc.