Warning: This project is deprecated. TensorFlow Addons has stopped development,
The project will only be providing minimal maintenance releases until May 2024. See the full
announcement here or on
github.
tfa.losses.NpairsLoss
Stay organized with collections
Save and categorize content based on your preferences.
Computes the npairs loss between y_true and y_pred.
tfa.losses.NpairsLoss(
name: str = 'npairs_loss'
)
Npairs loss expects paired data where a pair is composed of samples from
the same labels and each pairs in the minibatch have different labels.
The loss takes each row of the pair-wise similarity matrix, y_pred,
as logits and the remapped multi-class labels, y_true, as labels.
The similarity matrix y_pred between two embedding matrices a and b
with shape [batch_size, hidden_size] can be computed as follows:
a = tf.constant([[1, 2],
[3, 4],
[5, 6]], dtype=tf.float16)
b = tf.constant([[5, 9],
[3, 6],
[1, 8]], dtype=tf.float16)
y_pred = tf.matmul(a, b, transpose_a=False, transpose_b=True)
y_pred
<tf.Tensor: shape=(3, 3), dtype=float16, numpy=
array([[23., 15., 17.],
[51., 33., 35.],
[79., 51., 53.]], dtype=float16)>
<... Note: constants a & b have been used purely for
example purposes and have no significant value ...>
See: http://www.nec-labs.com/uploads/images/Department-Images/MediaAnalytics/papers/nips16_npairmetriclearning.pdf
Args |
|---|
name
|
(Optional) name for the loss.
|
Methods
from_config
@classmethod
from_config(
config
)
Instantiates a Loss from its config (output of get_config()).
| Args |
|---|
config
|
Output of get_config().
|
get_config
get_config()
Returns the config dictionary for a Loss instance.
__call__
__call__(
y_true, y_pred, sample_weight=None
)
Invokes the Loss instance.
| Args |
|---|
y_true
|
Ground truth values. shape = [batch_size, d0, .. dN], except
sparse loss functions such as sparse categorical crossentropy where
shape = [batch_size, d0, .. dN-1]
|
y_pred
|
The predicted values. shape = [batch_size, d0, .. dN]
|
sample_weight
|
Optional sample_weight acts as a coefficient for the
loss. If a scalar is provided, then the loss is simply scaled by the
given value. If sample_weight is a tensor of size [batch_size],
then the total loss for each sample of the batch is rescaled by the
corresponding element in the sample_weight vector. If the shape of
sample_weight is [batch_size, d0, .. dN-1] (or can be
broadcasted to this shape), then each loss element of y_pred is
scaled by the corresponding value of sample_weight. (Note
ondN-1: all loss functions reduce by 1 dimension, usually
axis=-1.)
|
| Returns |
|---|
Weighted loss float Tensor. If reduction is NONE, this has
shape [batch_size, d0, .. dN-1]; otherwise, it is scalar. (Note
dN-1 because all loss functions reduce by 1 dimension, usually
axis=-1.)
|
| Raises |
|---|
ValueError
|
If the shape of sample_weight is invalid.
|
Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License. For details, see the Google Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.
Last updated 2023-05-25 UTC.
[null,null,["Last updated 2023-05-25 UTC."],[],[],null,["# tfa.losses.NpairsLoss\n\n\u003cbr /\u003e\n\n|-------------------------------------------------------------------------------------------------------------------------|\n| [View source on GitHub](https://github.com/tensorflow/addons/blob/v0.20.0/tensorflow_addons/losses/npairs.py#L152-L191) |\n\nComputes the npairs loss between `y_true` and `y_pred`. \n\n tfa.losses.NpairsLoss(\n name: str = 'npairs_loss'\n )\n\nNpairs loss expects paired data where a pair is composed of samples from\nthe same labels and each pairs in the minibatch have different labels.\nThe loss takes each row of the pair-wise similarity matrix, `y_pred`,\nas logits and the remapped multi-class labels, `y_true`, as labels.\n\nThe similarity matrix `y_pred` between two embedding matrices `a` and `b`\nwith shape `[batch_size, hidden_size]` can be computed as follows: \n\n a = tf.constant([[1, 2],\n [3, 4],\n [5, 6]], dtype=tf.float16)\n b = tf.constant([[5, 9],\n [3, 6],\n [1, 8]], dtype=tf.float16)\n y_pred = tf.matmul(a, b, transpose_a=False, transpose_b=True)\n y_pred\n \u003ctf.Tensor: shape=(3, 3), dtype=float16, numpy=\n array([[23., 15., 17.],\n [51., 33., 35.],\n [79., 51., 53.]], dtype=float16)\u003e\n\n\\\u003c... Note: constants a \\& b have been used purely for\nexample purposes and have no significant value ...\\\u003e\n\nSee: \u003chttp://www.nec-labs.com/uploads/images/Department-Images/MediaAnalytics/papers/nips16_npairmetriclearning.pdf\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n| Args ---- ||\n|--------|-------------------------------|\n| `name` | (Optional) name for the loss. |\n\n\u003cbr /\u003e\n\nMethods\n-------\n\n### `from_config`\n\n @classmethod\n from_config(\n config\n )\n\nInstantiates a `Loss` from its config (output of `get_config()`).\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n| Args ||\n|----------|---------------------------|\n| `config` | Output of `get_config()`. |\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n| Returns ||\n|---|---|\n| A `Loss` instance. ||\n\n\u003cbr /\u003e\n\n### `get_config`\n\n get_config()\n\nReturns the config dictionary for a `Loss` instance.\n\n### `__call__`\n\n __call__(\n y_true, y_pred, sample_weight=None\n )\n\nInvokes the `Loss` instance.\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n| Args ||\n|-----------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|\n| `y_true` | Ground truth values. shape = `[batch_size, d0, .. dN]`, except sparse loss functions such as sparse categorical crossentropy where shape = `[batch_size, d0, .. dN-1]` |\n| `y_pred` | The predicted values. shape = `[batch_size, d0, .. dN]` |\n| `sample_weight` | Optional `sample_weight` acts as a coefficient for the loss. If a scalar is provided, then the loss is simply scaled by the given value. If `sample_weight` is a tensor of size `[batch_size]`, then the total loss for each sample of the batch is rescaled by the corresponding element in the `sample_weight` vector. If the shape of `sample_weight` is `[batch_size, d0, .. dN-1]` (or can be broadcasted to this shape), then each loss element of `y_pred` is scaled by the corresponding value of `sample_weight`. (Note on`dN-1`: all loss functions reduce by 1 dimension, usually axis=-1.) |\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n| Returns ||\n|---|---|\n| Weighted loss float `Tensor`. If `reduction` is `NONE`, this has shape `[batch_size, d0, .. dN-1]`; otherwise, it is scalar. (Note `dN-1` because all loss functions reduce by 1 dimension, usually axis=-1.) ||\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n| Raises ||\n|--------------|---------------------------------------------|\n| `ValueError` | If the shape of `sample_weight` is invalid. |\n\n\u003cbr /\u003e"]]
View source on GitHub