tensorflow
Repos
107

An Open Source Machine Learning Framework for Everyone

169561
70666

TensorFlow documentation

5532
4890

Stores documents used by the TensorFlow developer community

1135
513

Events

closed issue
Descriptions of `tf.RaggedTensor`'s methods '__xx__' are unclear

Documentation bug for: https://www.tensorflow.org/api_docs/python/tf/RaggedTensor?hl=en#and

Description of issue:

Usage examples are unmatched

tf.RaggedTensor has a number of class methods, for example, __abs__, __add__, ... The usage examples for these methods are unmatched. For instance, the example code for __add__ does not contain any ragged tensors, instead the examples are all normal tensors (of type tf.Tensor).

The example code is:

a = tf.constant([True])
b = tf.constant([False])
tf.math.logical_and(a, b)

That is not expected for a tf.RaggedTensor.__add__() method.

Created at 19 seconds ago
issue comment
Descriptions of `tf.RaggedTensor`'s methods '__xx__' are unclear

Closing as stale. Please reopen if you'd like to work on this further.

Created at 20 seconds ago
Created at 10 minutes ago
Created at 19 minutes ago
issue comment
Error in models defined under research folder -> "object_detection.protos.StringIntLabelMap" has 'no field named node'.

Did anyone find the solution? or workaround? Any help is appreciated. Thank you.

Created at 25 minutes ago
issue comment
Debugger V2 not working. Invalid argument: DebugNumericSummaryV2Op requires tensor_id to be less than or equal to (2^53). Given tensor_id:26

@jainmilan Could you try to use calling tf.config.run_functions_eagerly(True) that will make all invocations of tf.function run eagerly instead of running as a traced graph function. This can be useful for debugging. Please refer to this doc and let us know if it helps? Thank you!

Created at 32 minutes ago
closed issue
Support more stats_options through config.

The current (0.13.0) implementation of the StatisticsGen component only uses the default stats_options: stats_options = tensorflow_data_validation.statistics.stats_options.StatsOptions() There is a TODO statement saying that this should be configurable. There should be a default to the current behavior.

Created at 34 minutes ago
issue comment
Support more stats_options through config.

Hi, @martin-laurent

Closing this issue as it has been inactive for couple of weeks and Please feel free to reopen if this issue still persists. Thank you!

Created at 34 minutes ago
issue comment
Support the quantization of fused custom ops

"support string custom attributes" split as seperate PR: https://github.com/tensorflow/tensorflow/pull/58818.

Created at 36 minutes ago
pull request opened
Support string custom attributes

convert composite op to a TFLite custom operation , custom attributes could be set for custom operations through experimental_implements:

def get_implements_signature():
  implements_signature = [
    # 'name' will be used as a name for the operation.
    'name: "my_custom_fused_op"',
    # attr "tfl_fusable_op" is required to be set with true value.
    'attr {key: "tfl_fusable_op" value { b: true } }',
    # Example attribute "example_option" that the op accepts.
    'attr {key: "example_option" value { i: %d } }' % 10
  ]
  return ' '.join(implements_signature)

@tf.function(experimental_implements=get_implements_signature())
def my_custom_fused_op(input_1, input_2):

With this pull request, we enable string as supported data type of custom attributes:

def get_implements_signature():
    implements_signature = [
        ...
        # Example string attibute "custom_attr" that the op accepts
        'attr {key: "custom_attr" value { s: "value 1.01" } }',
        ...
    ]
    return " ".join(implements_signature)

@tf.function(experimental_implements=get_implements_signature())
def my_custom_fused_op(input_1, input_2):
Created at 42 minutes ago
started
Created at 43 minutes ago
issue comment
Parallel loading of nested ExtensionTypes produces shape mismatch

This could have a common root cause with https://github.com/tensorflow/tensorflow/issues/58676 and https://github.com/tensorflow/tensorflow/issues/58140 (TF_RUN_EAGER_OP_AS_FUNCTION env variable).

Created at 45 minutes ago
issue comment
tf.keras.Model with multiple outputs, using fit() with a generator dataset passes y to loss with wrong shape

Hi @mdatsev , I have done the necessary changes in the code and now the model is able to train with the Generator Dataset also.Also please note that you have to apply break condition also to exit the loop which was missing in the earlier code.Please refer the attached gist.

Created at 45 minutes ago
pull request opened
Bump decode-uri-component from 0.2.0 to 0.2.2 in /react-native/image-classification/react-native-cli

Bumps decode-uri-component from 0.2.0 to 0.2.2.

Dependabot compatibility score

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.


You can trigger Dependabot actions by commenting on this PR:

  • @dependabot rebase will rebase this PR
  • @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
  • @dependabot merge will merge this PR after your CI passes on it
  • @dependabot squash and merge will squash and merge this PR after your CI passes on it
  • @dependabot cancel merge will cancel a previously requested merge and block automerging
  • @dependabot reopen will reopen this PR if it is closed
  • @dependabot close will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually
  • @dependabot ignore this major version will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)
  • @dependabot ignore this minor version will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)
  • @dependabot ignore this dependency will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself) You can disable automated security fix PRs for this repo from the Security Alerts page.
Created at 45 minutes ago
create branch
dependabot[bot] create branch dependabot/npm_and_yarn/react-native/image-classification/react-native-cli/decode-uri-component-0.2.2
Created at 45 minutes ago
issue comment
Use cl_khr_integer_dot_product if available

Sure, the idea is to allow TFLite's GPU delegate to make use of those instructions in GPGPUs that optimize the dot product of integers:

https://registry.khronos.org/OpenCL/specs/3.0-unified/html/OpenCL_Ext.html#_new_opencl_c_functions_2

Created at 46 minutes ago
Created at 46 minutes ago
issue comment
BeamDagRunner cannot run BigQueryExampleGen

Are you satisfied with the resolution of your issue? Yes No

Created at 52 minutes ago
closed issue
BeamDagRunner cannot run BigQueryExampleGen

Repro steps:

  • Create a simple pipeline with only one component: BigQueryExampleGen
  • Instantiate a BeamDagRunner and use it to run the example gen component
  • Observe error: "unhashable type: 'Environment' [while running 'Run[BigQueryExampleGen]']"

Colab where I repro this: https://colab.research.google.com/drive/1Nj-BJRS6zFm4w1kXffKyh31VLWpBTHeJ

Note that you will not be able to run the codelab as you don't have access to the credentials file. But the error should be visible.

Created at 52 minutes ago

A ranking environment that explicitly models observation probabilities.

PiperOrigin-RevId: 493826467 Change-Id: Id3c2c7007a69d2b47eeb2ccc60fae1ca66b24030

Created at 52 minutes ago
issue comment
BeamDagRunner cannot run BigQueryExampleGen

Hi, @misterpeddy

Closing this issue as it has been inactive for couple of weeks and Please feel free to reopen if this issue still persists. Thank you!

Created at 53 minutes ago
issue comment
Updated ML Pipeline Link

I think that we can just close this PR. :)

Created at 56 minutes ago

Do not create softmax fusion for f64 type.

Our codegen cannot support reductions with f64 type.

PiperOrigin-RevId: 493823693

Created at 57 minutes ago
create branch
copybara-service[bot] create branch test_493817924
Created at 57 minutes ago
pull request opened
Move imagenet2012_corrupted to 'datasets/' directory, using ConfigBasedBuilder.

Move imagenet2012_corrupted to 'datasets/' directory, using ConfigBasedBuilder.

Created at 57 minutes ago
issue comment
[2.2] XLA requires 2x GPU memory with sparse_categorical_crossentropy

This issue has been automatically marked as stale because it has no recent activity. It will be closed if no further activity occurs. Thank you.

Created at 1 hour ago

compat: Update forward compatibility horizon to 2022-12-08

PiperOrigin-RevId: 493823072

Created at 1 hour ago