tf.debugging.assert_near
Stay organized with collections
Save and categorize content based on your preferences.
Assert the condition x and y are close element-wise.
tf.debugging.assert_near(
x, y, rtol=None, atol=None, data=None, summarize=None, message=None, name=None
)
Example of adding a dependency to an operation:
with tf.control_dependencies([tf.compat.v1.assert_near(x, y)]):
output = tf.reduce_sum(x)
This condition holds if for every pair of (possibly broadcast) elements
x[i], y[i], we have
If both x and y are empty, this is trivially satisfied.
The default atol and rtol is 10 * eps, where eps is the smallest
representable positive number such that 1 + eps != 1. This is about
1.2e-6 in 32bit, 2.22e-15 in 64bit, and 0.00977 in 16bit.
See numpy.finfo.
Args |
|---|
x
|
Float or complex Tensor.
|
y
|
Float or complex Tensor, same dtype as, and broadcastable to, x.
|
rtol
|
Tensor. Same dtype as, and broadcastable to, x.
The relative tolerance. Default is 10 * eps.
|
atol
|
Tensor. Same dtype as, and broadcastable to, x.
The absolute tolerance. Default is 10 * eps.
|
data
|
The tensors to print out if the condition is False. Defaults to
error message and first few entries of x, y.
|
summarize
|
Print this many entries of each tensor.
|
message
|
A string to prefix to the default message.
|
name
|
A name for this operation (optional). Defaults to "assert_near".
|
Returns |
|---|
Op that raises InvalidArgumentError if x and y are not close enough.
|
Numpy Compatibility
Similar to numpy.assert_allclose, except tolerance depends on data type.
This is due to the fact that TensorFlow is often used with 32bit, 64bit,
and even 16bit data.
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 2020-10-01 UTC.
[null,null,["Last updated 2020-10-01 UTC."],[],[],null,["# tf.debugging.assert_near\n\n\u003cbr /\u003e\n\n|-------------------------------------------------------------------|-----------------------------------------------------------------------------------------------------------------------------|\n| [TensorFlow 2 version](/api_docs/python/tf/debugging/assert_near) | [View source on GitHub](https://github.com/tensorflow/tensorflow/blob/v1.15.0/tensorflow/python/ops/check_ops.py#L761-L837) |\n\nAssert the condition `x` and `y` are close element-wise.\n\n#### View aliases\n\n\n**Main aliases**\n\n\\`tf.assert_near\\`\n**Compat aliases for migration**\n\nSee\n[Migration guide](https://www.tensorflow.org/guide/migrate) for\nmore details.\n\n[`tf.compat.v1.assert_near`](/api_docs/python/tf/compat/v1/assert_near), [`tf.compat.v1.debugging.assert_near`](/api_docs/python/tf/compat/v1/assert_near)\n\n\u003cbr /\u003e\n\n tf.debugging.assert_near(\n x, y, rtol=None, atol=None, data=None, summarize=None, message=None, name=None\n )\n\nExample of adding a dependency to an operation: \n\n with tf.control_dependencies([tf.compat.v1.assert_near(x, y)]):\n output = tf.reduce_sum(x)\n\nThis condition holds if for every pair of (possibly broadcast) elements\n`x[i]`, `y[i]`, we have \n\n```bash\n\n```\n\nIf both `x` and `y` are empty, this is trivially satisfied.\n\nThe default `atol` and `rtol` is `10 * eps`, where `eps` is the smallest\nrepresentable positive number such that `1 + eps != 1`. This is about\n`1.2e-6` in `32bit`, `2.22e-15` in `64bit`, and `0.00977` in `16bit`.\nSee `numpy.finfo`.\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n| Args ---- ||\n|-------------|------------------------------------------------------------------------------------------------------------------|\n| `x` | Float or complex `Tensor`. |\n| `y` | Float or complex `Tensor`, same `dtype` as, and broadcastable to, `x`. |\n| `rtol` | `Tensor`. Same `dtype` as, and broadcastable to, `x`. The relative tolerance. Default is `10 * eps`. |\n| `atol` | `Tensor`. Same `dtype` as, and broadcastable to, `x`. The absolute tolerance. Default is `10 * eps`. |\n| `data` | The tensors to print out if the condition is False. Defaults to error message and first few entries of `x`, `y`. |\n| `summarize` | Print this many entries of each tensor. |\n| `message` | A string to prefix to the default message. |\n| `name` | A name for this operation (optional). Defaults to \"assert_near\". |\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n| Returns ------- ||\n|---|---|\n| Op that raises `InvalidArgumentError` if `x` and `y` are not close enough. ||\n\n\u003cbr /\u003e\n\n#### Numpy Compatibility\n\nSimilar to `numpy.assert_allclose`, except tolerance depends on data type.\nThis is due to the fact that `TensorFlow` is often used with `32bit`, `64bit`,\nand even `16bit` data."]]
TensorFlow 2 version
View source on GitHub