Weekly GitHub Report for Tensorflow: July 14, 2025 - July 21, 2025 (12:08:37)

Weekly GitHub Report for Tensorflow

Thank you for subscribing to our weekly newsletter! Each week, we deliver a comprehensive summary of your GitHub project's latest activity right to your inbox, including an overview of your project's issues, pull requests, contributors, and commit activity.

---

Table of Contents

• I. News
  • 1.1. Recent Version Releases
  • 1.2. Other Noteworthy Updates
• II. Issues
  • 2.1. Top 5 Active Issues
  • 2.2. Top 5 Stale Issues
  • 2.3. Open Issues
  • 2.4. Closed Issues
  • 2.5. Issue Discussion Insights

• III. Pull Requests
  • 3.1. Open Pull Requests
  • 3.2. Closed Pull Requests
  • 3.3. Pull Request Discussion Insights

• IV. Contributors
  • 4.1. Contributors

I. News

1.1 Recent Version Releases:

The current version of this repository is v2.19.0

1.2 Version Information:

The TensorFlow 2.19.0 release, created on March 5, 2025, introduces breaking changes to the LiteRT C++ and Python APIs, including the transition of tf.lite.Interpreter to a new location with a deprecation warning. Notable improvements include support for bfloat16 in the tfl.Cast operation, and the discontinuation of publishing libtensorflow packages, which can still be accessed via PyPI.

II. Issues

2.1 Top 5 Active Issues:

We consider active issues to be issues that that have been commented on most frequently within the last week. Bot comments are omitted.

1. Inconsistent behavior for tf.raw_ops.NotEqual between CPU and GPU with non-broadcastable shapes: This issue highlights a bug in TensorFlow where the tf.raw_ops.NotEqual operation behaves inconsistently between CPU and GPU when dealing with non-broadcastable shapes. The GPU correctly raises an InvalidArgumentError for invalid inputs, while the CPU silently returns a misleading scalar value, violating the principle of device consistency.

   • A contributor expressed interest in resolving the issue by updating the CPU implementation to match the GPU's error handling. Another participant encouraged the contributor to proceed without needing permission and provided guidance on where to focus the changes in the codebase. The contributor acknowledged the advice and planned to address the kernel issue, ensuring consistent error handling across devices, and intended to submit a pull request once the fix and tests were ready.
   • Number of comments this week: 3

2. Tensorflow 2.19 fails to load after Pyside6: This issue describes a bug where TensorFlow 2.19.0 fails to load after importing PySide6 on a Windows 11 system with Python 3.10.8. The error occurs specifically when the import order is reversed, leading to a DLL load failure related to TensorFlow's native runtime.

   • The comments discuss potential causes for the issue, such as missing MSVC 2019 redistributable, lack of AVX2 support, or a 32-bit CPU/Python installation. One comment suggests the problem might be an ODR violation rather than a duplicate of a previously reported issue.
   • Number of comments this week: 2

3. could you add support of the new optimizer: Muon: This issue is about a request to add support for a new optimizer called Muon in TensorFlow version 2.19, as the current version lacks this feature. The user has confirmed that the issue persists even with TensorFlow Nightly and has provided a link to the Muon optimizer for reference.

   • The comments discuss a request for the user to submit a pull request (PR) to add the Muon optimizer support. The user has submitted a PR but mentions that they are inexperienced and may need assistance from the maintainers to finalize it.
   • Number of comments this week: 2

4. tf.experimental.numpy.cumsum handles overflow inconsistently on CPU and GPU: This issue highlights a bug in TensorFlow where the tf.experimental.numpy.cumsum function handles overflow inconsistently between CPU and GPU when using the int16 data type. The CPU version aligns with the behavior of NumPy, while the GPU version does not, resulting in different outputs for the same operation.

   • A user confirmed the issue by reproducing it with TensorFlow versions 2.19.0 and 2.18.0, providing a Colab gist for reference.
   • Number of comments this week: 1

Since there were fewer than 5 open issues, all of the open issues have been listed above.

2.2 Top 5 Stale Issues:

We consider stale issues to be issues that has had no activity within the last 30 days. The team should work together to get these issues resolved and closed as soon as possible.

1. TF-TRT Warning: Could not find TensorRT: This issue involves a user experiencing difficulties with TensorFlow on an Ubuntu 22.04 system, specifically encountering a warning that TensorRT could not be found despite multiple attempts to install and configure the software within an Anaconda environment. The user suspects that the problem may be related to driver compatibility issues with their NVIDIA RTX 3050 TI GPU, as they are unable to use the recommended 550 driver and must revert to the 535 driver, which may not fully support CUDA 12.4.
2. SystemError in tf.ensure_shape and tf.compat.v1.ensure_shape when dtype of shape is tf.uint64 and its value is too large.: This issue involves a bug in TensorFlow where using tf.ensure_shape or tf.compat.v1.ensure_shape with a shape of dtype tf.uint64 and a value close to 2^64 results in a SystemError and OverflowError. The problem occurs specifically when the shape is defined with large values, such as tf.constant([18446743219011059112, 1], dtype=tf.uint64), causing these errors to be triggered in eager execution mode.
3. Feature Request: Integrate different Digital Signal Processing into tf.signal: This issue is a feature request to integrate advanced Digital Signal Processing (DSP) functionalities into TensorFlow's tf.signal module, similar to those available in the PyTorch ecosystem, particularly from the julius library. The integration aims to enhance TensorFlow's capabilities in audio data augmentation, providing researchers and developers with native tools for complex audio processing, thereby improving workflow efficiency and model generalization without relying on external libraries.
4. [DOCS] Missing complex input for Round op: This issue highlights a documentation bug in TensorFlow's Round operation, where the official documentation incorrectly states that a complex tensor can be directly used as input for the operation. The user reports that attempting to use a complex tensor results in an error, and they must separately apply the Round operation to the real and imaginary parts of the tensor to achieve the expected outcome.
5. tf.raw_ops.Unbatch aborts with "Check failed: d < dims()": This issue reports a bug in TensorFlow version 2.17 where the tf.raw_ops.Unbatch operation aborts unexpectedly with an error message indicating a failed check on tensor dimensions. The problem occurs on a Linux Ubuntu 20.04.3 LTS system using Python 3.11.8, and the error can be reproduced with TensorFlow Nightly, suggesting a persistent issue in the codebase.

2.3 Open Issues

This section lists, groups, and then summarizes issues that were created within the last week in the repository.

Issues Opened This Week: 9

Summarized Issues:

• Inconsistent Behavior Between CPU and GPU: Several issues highlight inconsistencies in TensorFlow's behavior between CPU and GPU executions. These include the tf.experimental.numpy.cumsum function with int16 data type, tf.linalg.matrix_rank with a tolerance of 6, tf.pow with int64 tensors, and tf.raw_ops.NotEqual with non-broadcastable shapes. In each case, one device aligns with expected results (often matching NumPy), while the other does not, leading to potential errors and unexpected outputs.
  • issues/97042, issues/97102, issues/97125, issues/97204

• Function Errors and Crashes: TensorFlow has encountered several function-specific errors and crashes. The tf.math.argmax function throws an InvalidArgumentError with certain tensor types, and tf.nn.depthwise_conv2d crashes with large strides values on CPU when ONEDNN is enabled. These issues highlight the need for improved error handling and compatibility checks within TensorFlow's functions.
  • issues/97096, issues/97165

• Compatibility and Dependency Issues: TensorFlow 2.19.0 fails to load after importing PySide6 on Windows 11, resulting in a DLL load failure error. This issue suggests potential problems with missing dependencies or an ODR violation, indicating a need for better compatibility and dependency management in TensorFlow's installation process.
  • issues/97058

• Incorrect Function Outputs: The tf.nn.local_response_normalization function produces incorrect outputs on CPU with a negative beta value and encounters errors on GPU due to cuDNN's requirements. This discrepancy with PyTorch's results suggests a need for TensorFlow to align its function outputs with expected standards across different platforms.
  • issues/97105

• Feature Request for New Optimizer: There is a request to add support for a new optimizer called Muon in TensorFlow version 2.19. A contributor has already submitted a pull request and is seeking assistance from maintainers, indicating community interest in expanding TensorFlow's optimization capabilities.
  • issues/97187

2.4 Closed Issues

This section lists, groups, and then summarizes issues that were closed within the last week in the repository. This section also links the associated pull requests if applicable.

Issues Closed This Week: 2

Summarized Issues:

• Import Failure with Optimization Flag: A bug in TensorFlow version 2.19.0 causes import failures when Python is run with the -OO optimization flag. This issue arises because the code in tensorflow/python/util/all_util.py depends on docstrings, which are removed at this optimization level, preventing programs from running at the highest optimization level.
  • issues/96900

• Inconsistent Error Handling in tf.raw_ops.NotEqual: TensorFlow 2.19.0 exhibits inconsistent behavior in the tf.raw_ops.NotEqual operation between CPU and GPU. The CPU implementation fails silently and returns an incorrect scalar value for non-broadcastable tensor shapes, while the GPU correctly raises an InvalidArgumentError, highlighting the need for uniform error handling to avoid silent bugs.
  • issues/97227

2.5 Issue Discussion Insights

This section will analyze the tone and sentiment of discussions within this project's open and closed issues that occurred within the past week. It aims to identify potentially heated exchanges and to maintain a constructive project environment.

Based on our analysis, there are no instances of toxic discussions in the project's open or closed issues from the past week.

---

III. Pull Requests

3.1 Open Pull Requests

This section provides a summary of pull requests that were opened in the repository over the past week. The top three pull requests with the highest number of commits are highlighted as 'key' pull requests. Other pull requests are grouped based on similar characteristics for easier analysis. Up to 25 pull requests are displayed in this section, while any remaining pull requests beyond this limit are omitted for brevity.

Pull Requests Opened This Week: 3

Key Open Pull Requests

1. Change int64 to int64_t in Pow operations for type consistency: This pull request standardizes the integer type usage in the Pow operation registrations within TensorFlow by replacing int64 with int64_t for both CPU and GPU implementations to ensure type consistency, addressing issue #97125 and refining the implementation based on feedback from a previous related pull request.

• URL: pull/97210

• Merged: No

• Associated Commits: c95df, 18c00

2. lite: Add config option to enable benchmark_model: This pull request introduces a configuration option to enable the benchmark_model tool in TensorFlow Lite, allowing it to be built and installed using the "all" target, which is beneficial for certain packaging scenarios, and serves as the benchmark_model counterpart to a previous change referenced as #60021.

• URL: pull/96999

• Merged: No

• Associated Commits: c9a1c

3. Add a scratch implemention of muon: This pull request introduces a preliminary implementation of the "muon" feature to the TensorFlow project, acknowledging potential errors due to the author's limitations and suggesting that reviewers refer to the existing implementation in Keras for guidance and corrections.

• URL: pull/97231

• Merged: No

• Associated Commits: 77b01

3.2 Closed Pull Requests

This section provides a summary of pull requests that were closed in the repository over the past week. The top three pull requests with the highest number of commits are highlighted as 'key' pull requests. Other pull requests are grouped based on similar characteristics for easier analysis. Up to 25 pull requests are displayed in this section, while any remaining pull requests beyond this limit are omitted for brevity.

Pull Requests Closed This Week: 7

Key Closed Pull Requests

1. Feature/fcc: This pull request involves the development and debugging of a feature for TensorFlow that implements a fully connected delegate with support for FPGA, including the addition of various test delegates, build file updates, and checks for dynamic tensors, but it was ultimately not merged.

• URL: pull/97133

• Merged: No

• Associated Commits: a0b6d, 2bd92, e9d01, 0d853, c46fd, ddf47, dfae1, 3d304, b17be, d95a8, 93c0a, 84a4e, 8448f, 99cc4, 17d7a, 0a0ed, 2846a, 8e1e0, 4c8aa, 992e4, 4c246, 480d2, a3d38, 2176a, 5214a, 54c3a, baac0, ef8ad, eb716, 7bc1e, 16b7b, ec4c8, 38b33, ce4fc, a7fed, ab1a3, 087ef, db460, 2a84a, 7f757, 92593, 3dfb7, 959c7, 41b8d, b4188, 318b6, 4a2e2, d504e, 6a93f, 4e4d7, 333ab, fd0f6, 87707, 4cebf, d32bb, 20f28, dcb32, 9f467, 15696, 0acb4, 87ce2, a1cce, ce3c0, 84146, 7c4b5, a23f2, 61173, 236af, 77cd7, 4c900, db5c0, ddf63, 5f9ca, 94704, 8cc8f, 36fce, 7deec, 55e0a, da11c, 0ac37, 6e14f, 493ad, fd97d, d6bee, c7365, 5a9fc, ffa2a, cb733, 4dc33, 57c59, f0484, 0abd5, 280ee

2. Update release notes for TensorFlow 2.20.0: This pull request involves updating the release notes for TensorFlow version 2.20.0, where it was initially incomplete and required a Release Owner to fill in the internal release notes before final submission, as indicated by the commits that inserted a placeholder and subsequently updated the RELEASE.md file.

• URL: pull/97080

• Merged: Yes

• Associated Commits: 5a4d2, ae214

3. Solve the problem number #97125: This pull request attempts to address issue number #97125 by improving syntax and modifying GPU-related code from int64 to int64_t, but it was ultimately not merged.

• URL: pull/97131

• Merged: No

• Associated Commits: 45476, 99a47

Other Closed Pull Requests

• README.md Updates: Two pull requests focused on updating the README.md file in the TensorFlow project. One aimed to correct minor grammatical issues but was not merged, while the other successfully addressed and fixed these issues, resulting in a merged update.
  • pull/96898, pull/96902

• Docstring Absence Issue: A pull request addressed a problem in the TensorFlow project where the lack of docstrings, due to Python's optimization flag -OO, could lead to failures. This issue was successfully resolved and merged into the codebase.
  • pull/96906

• Flaky Test Attempts Option: A pull request enabled the --flaky_test_attempts option in the release branch of the TensorFlow project. The changes were successfully merged, as indicated by the commit message.
  • pull/97076

3.3 Pull Request Discussion Insights

This section will analyze the tone and sentiment of discussions within this project's open and closed pull requests that occurred within the past week. It aims to identify potentially heated exchanges and to maintain a constructive project environment.

Based on our analysis, there are no instances of toxic discussions in the project's open or closed pull requests from the past week.

---

IV. Contributors

4.1 Contributors

Active Contributors:

We consider an active contributor in this project to be any contributor who has made at least 1 commit, opened at least 1 issue, created at least 1 pull request, or made more than 2 comments in the last month.

If there are more than 10 active contributors, the list is truncated to the top 10 based on contribution metrics for better clarity.

Contributor	Commits	Pull Requests	Issues	Comments
harshithn31	93	1	0	0
CloudSmallInsect	0	0	8	27
Venkat6871	1	0	0	14
mihaimaruseac	0	0	0	14
dhruv-dhiman122	4	2	0	2
mynickmynick	6	1	0	0
jiren-the-gray	0	0	7	0
gaikwadrahul8	0	0	0	7
gzqccnu	1	1	2	1
Tai78641	3	1	0	0