Weekly GitHub Report for Tensorflow: October 20, 2025 - October 27, 2025 (12:03:26)

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:

Released on March 5, 2025, TensorFlow version 2.19.0 introduces breaking changes to the tf.lite API, including the deprecation of tf.lite.Interpreter in favor of ai_edge_litert.interpreter and changes to certain C++ constants for better API compatibility. Key updates also include runtime support for the bfloat16 data type in the tfl.Cast operation, alongside the discontinuation of standalone libtensorflow package publishing, while still allowing unpacking from 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. GRU character generation functional API model throws various exceptions.: This issue reports that using a GRU layer with a functional API model in TensorFlow 2.20.0 causes various exceptions, particularly when providing an initial state to the GRU. The user experiences KeyError exceptions related to model outputs and loss paths during training, and the problem persists even when the initial state is omitted, indicating a deeper issue with model compilation or output handling.

   • The comment acknowledges the issue and informs the user that a community member has submitted a pull request intended to fix the problem; it also apologizes for the delay and expresses gratitude for reporting the issue, indicating that the fix will be available once the PR is merged.
   • Number of comments this week: 1

2. Division Precision Problem in Graph Mode on Intel CPU: This issue reports a precision problem in TensorFlow's graph mode division operation on Intel and AMD CPUs starting from version 2.9, where dividing tensors of ones results in values slightly less than 1.0 instead of exactly 1.0. The problem does not occur in eager mode or on Apple M4 CPUs, and it persists regardless of enabling or disabling oneDNN optimizations, with Eigen's cwiseQuotient providing accurate results under AVX2 or AVX512F.

   • The single comment reproduces the bug with provided code and logs, confirming the precision discrepancy occurs only in graph mode on specific CPUs and TensorFlow versions, and clarifies that the issue is not related to oneDNN settings or present in earlier TensorFlow releases.
   • Number of comments this week: 1

3. Memory usage stop increasing when max_range increase in func tf.raw_ops.QuantizeV2(): This issue reports a potential bug in the TensorFlow function tf.raw_ops.QuantizeV2(), where GPU memory usage does not consistently increase as the max_range parameter is increased, contrary to expectations based on initial tests. The user provides multiple reproducible code examples showing that memory usage rises when max_range changes from 100 to 101 but then remains constant even when max_range is set to a very large value, questioning whether this behavior is intended or due to some internal optimization.

   • A commenter expressed interest in investigating the issue and asked for clarification on why memory usage is expected to increase with max_range, seeking to understand the underlying cause of the observed behavior.
   • 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 describes a problem where the user is encountering a warning that TensorRT cannot be found when running TensorFlow 2.15.0 on an Ubuntu 22.04 system with an RTX 3050 Ti GPU, despite having installed CUDA 12.4 and using the NVIDIA 535 driver. The user reports difficulties with driver compatibility and has tried reinstalling TensorFlow multiple times, but continues to receive the TF-TRT warning, seeking assistance to resolve the TensorRT detection failure.
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 reports a bug in TensorFlow where calling tf.ensure_shape or tf.compat.v1.ensure_shape with a shape tensor of type tf.uint64 containing very large values close to 2^64 triggers a SystemError and OverflowError. The problem occurs specifically in eager execution mode, causing these APIs to fail when handling such large shape values, which is unexpected behavior.
3. Feature Request: Integrate different Digital Signal Processing into tf.signal: This issue is a feature request proposing the integration of advanced digital signal processing (DSP) functionalities, similar to those found in the PyTorch julius library, into TensorFlow's tf.signal module. The requester highlights the benefits of having native audio data augmentation tools within TensorFlow to improve model robustness and streamline workflows by reducing reliance on external libraries.
4. [DOCS] Missing complex input for Round op: This issue reports a documentation bug concerning the Round operation in TensorFlow, where the official documentation states that complex tensors are supported as input, but in practice, attempting to use a complex tensor with tf.raw_ops.Round results in an error. The user found that the operation does not work as expected on complex inputs and must instead apply the rounding separately to the real and imaginary parts, indicating a discrepancy between the documented behavior and the actual implementation.
5. tf.raw_ops.Unbatch aborts with "Check failed: d < dims()": This issue reports a bug in TensorFlow version 2.17 where the operation tf.raw_ops.Unbatch aborts with a fatal check failure error "Check failed: d < dims()" when invoked with certain random tensor inputs. The problem occurs on Linux Ubuntu 20.04.3 LTS using Python 3.11.8, and the user has provided a minimal reproducible code snippet demonstrating the crash, which results in the program aborting with a core dump.

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: 5

Summarized Issues:

• Build and Compilation Errors: Multiple compilation errors occur when building TensorFlow Lite for Android with specific CMake options, particularly in the xnnpack delegate's file_util.cc file. These errors involve undefined types and mismatched declarations related to the FileDescriptor and FileDescriptorView classes, preventing successful compilation.
• issues/102572

• Model Training and API Bugs: Using a GRU layer with an initial state in a TensorFlow functional API model causes KeyError exceptions during training due to mismatches in model inputs, outputs, and loss configuration. This bug disrupts character generation model workflows and highlights issues in handling initial states within the functional API.
• issues/102598

• Version Compatibility Issues: TensorFlow 2.14 built against NumPy 1.x fails to operate correctly when installed alongside NumPy 2.3.4 or higher, leading to runtime errors. The issue raises concerns about why version constraints do not prevent incompatible combinations of TensorFlow and NumPy versions.
• issues/102743

• Precision and Numerical Accuracy Bugs: TensorFlow 2.9 exhibits a precision bug where division of float32 tensors in graph mode on certain Intel and AMD CPUs results in slightly inaccurate values just under 1.0, unlike earlier versions or Apple M4 CPUs. This issue is unaffected by oneDNN optimizations and indicates CPU-specific numerical discrepancies.
• issues/102771

• GPU Memory Usage Anomalies: The tf.raw_ops.QuantizeV2() function shows unexpected GPU memory usage patterns where increasing the max_range parameter slightly raises memory use, but setting it to a very large value stops the increase. This behavior may be due to an internal optimization or a potential bug in memory management.
• issues/102812

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: 3

Summarized Issues:

• Spam submissions: Multiple issues report spam submissions titled "I'm a spammer" that were identified and removed from the TensorFlow GitHub repository. These spam reports indicate attempts to misuse the issue tracker for irrelevant content, which were promptly addressed by the maintainers.
• issues/102778, issues/102799

• Library alignment problem: An issue highlights that the native library arm64-v8a/libtask_text_jni.so from the org.tensorflow:tensorflow-lite-task-text:0.4.4 package is not aligned to the required 16 KB boundary. This misalignment could cause runtime problems or inefficiencies when using the library on compatible devices.
• issues/102566

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: 4

Key Open Pull Requests

1. Add support for Linux RISC-V 64: This pull request adds support for compiling TensorFlow on Linux RISC-V 64 by fixing build system issues, updating dependencies, and including necessary patches and tests to enable proper functionality on this architecture.

• URL: pull/102607

• Merged: No

• Associated Commits: 6e38c, 3f2b8, 7c93b, d9a37, 184a4, 0ebb5

2. Fix/tflite reduce quantized get tensor data t: This pull request addresses an overflow issue in the TensorFlow Lite ReduceSum reference kernel by replacing the use of uint8_t with a templated GetTensorData method and includes a test to verify the fix.

• URL: pull/102775

• Merged: No

• Associated Commits: 8f8a8, 94cb1

3. Fix TensorFlow Lite build error with Android NDK (#102572) : This pull request addresses a build error in TensorFlow Lite when using the Android NDK by adding a platform-specific guard to conditionally include the header only on non-Windows systems, thereby preventing compilation failures caused by the unconditional inclusion of this header on Windows and certain Android NDK configurations.

• URL: pull/102674

• Merged: No

• Associated Commits: ef974

Other Open Pull Requests

• TPU Input Placement Enhancements: This pull request focuses on enabling the placement of TPU input on corresponding local CPU devices to optimize performance. The change is applied by cherrypicking a specific commit into the r2.19 branch of the TensorFlow project, ensuring compatibility with that release line.
• pull/102842

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: 3

Key Closed Pull Requests

1. Fix TypeError when serializing tf.nn.log_softmax activation: This pull request fixes a TypeError encountered during the deserialization of Keras models using the tf.nn.log_softmax activation by mapping the internal TensorFlow 2.x function name 'log_softmax_v2' to its canonical name 'log_softmax' in the _TF_ACTIVATIONS_V2 dictionary, ensuring proper serialization and deserialization compatibility.

• URL: pull/102663

• Merged: Yes

• Associated Commits: a1dab

2. I'm a shameless spammer: This pull request addresses a typo correction in the contribution guidelines of the TensorFlow project but was not merged.

• URL: pull/102722

• Merged: No

• Associated Commits: fb234

3. Fix GRU functional API model with return_state=True throwing KeyError: This pull request addresses a KeyError issue in TensorFlow's Functional API when using GRU or other RNN layers with return_state=True by adding handling in map_to_output_names() to correctly unwrap single-element nested loss and metric structures for models with a single output.

• URL: pull/102791

• Merged: No

• Associated Commits: f084c

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
ILCSFNO	13	8	10	6
CodersAcademy006	11	10	0	9
kshiteej-mali	7	1	0	0
Venkat6871	1	1	0	6
infiWang	6	1	0	1
codingWizard-Nikhil	5	2	0	0
souhil25	4	2	0	1
jameslovespancakes	3	3	0	0
khteh	0	0	5	1
syan427	0	0	6	0