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 TensorFlow Lite (tf.lite) API, including the deprecation of tf.lite.Interpreter in favor of ai_edge_litert.interpreter and changes to certain C++ constants for improved API flexibility. 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. [TYPE:BUG] Missing depthwise conv 16x8 xtensa kernel: This issue addresses the lack of an optimized depthwise convolution kernel for the Xtensa platform when using 16-bit activations and 8-bit weights, causing the system to default to a less efficient reference kernel instead. The user highlights the need for leveraging the existing optimized implementation to improve performance on Xtensa devices.

   • The comments indicate that the issue was misplaced and has been transferred to the TensorFlow Lite Micro repository, with one user confirming resolution and another discussing compliance requirements related to kernel specialization under the EU AI Act, ultimately suggesting the issue can be closed.
   • Number of comments this week: 4

2. [TYPE:BUG] [COMP:XLA] [TF 2.19] [XLA] Heap corruption (free(): invalid next size) when tf.slice produces zero-sized output from dynamically-shaped input: This issue describes a bug where using tf.slice to produce a zero-sized output from a dynamically-shaped input tensor inside a jit-compiled function causes a heap corruption error in TensorFlow's XLA mode, while eager mode handles the operation correctly. The problem occurs specifically when tf.slice is called with a zero-sized size argument on a dynamically-shaped tensor, leading to a crash with an invalid memory error during execution.

   • The single comment confirms the issue can be reproduced on Colab using TensorFlow versions 2.20.0 and nightly builds, providing a shared notebook for reference and verification.
   • Number of comments this week: 1

3. [TYPE:BUG] [COMP:OPS] [AWAITING PR MERGE] [TF 2.19] tf.linalg.diag aborts with CHECK failure (d >= 0) instead of raising InvalidArgumentError for invalid band k with rank-1 input: This issue reports that calling tf.linalg.diag with an invalid band specification on a rank-1 input tensor causes TensorFlow to abort with a C++ CHECK failure instead of raising a proper Python InvalidArgumentError. The problem occurs in both eager and XLA modes and has been reproduced on TensorFlow Nightly 2.21.0-dev, highlighting a need for improved error handling in this function.

   • The single comment confirms replication of the error on TensorFlow 2.21.0rc0 and tf-nightly in Google Colab, providing a link to a notebook demonstrating the issue.
   • Number of comments this week: 1

4. [TYPE:BUG] [STAT:AWAITING RESPONSE] [COMP:OPS] [TF 2.19] tf.raw_ops.AvgPool3DGrad aborts with CHECK failure in MKL path for invalid grad rank (d < dims()), instead of raising InvalidArgumentError: This issue reports that calling tf.raw_ops.AvgPool3DGrad with a non-5D gradient tensor in the NDHWC data format causes the Python process to abort due to a CHECK failure in the MKL path, rather than raising an InvalidArgumentError as expected. The user provides a minimal reproducible example demonstrating the crash and includes relevant log output showing the failure stack trace.

   • A commenter tested the provided code and was unable to reproduce the crash, instead encountering the expected InvalidArgumentError for passing a 2D tensor instead of a 5D tensor, and shared a Colab gist for further 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.

As of our latest update, there are no stale issues for the project this week.

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:

• XLA JIT Compilation Crashes: Multiple issues describe crashes and aborts occurring during XLA JIT compilation in TensorFlow when handling specific operations like tf.slice producing zero-sized outputs or using StackV2 with tf.linalg.qr. These crashes are caused by heap corruption or invalid dimension errors, leading to process aborts instead of graceful error handling.
• [issues/110789, issues/110798]

• Invalid Argument Handling and CHECK Failures: Several bugs involve TensorFlow aborting with C++ CHECK failures instead of raising proper Python exceptions when invalid arguments are passed, such as invalid band specifications in tf.linalg.diag or incorrect tensor shapes in tf.raw_ops.AvgPool3DGrad. This results in abrupt process termination rather than user-friendly error messages.
• [issues/110796, issues/110797]

• Documentation Issues with Keras Metrics: The TensorFlow documentation incorrectly advises using add_metric() inside the call() method of Keras models, which is broken in TensorFlow 2.17.0 with Keras 3.4.1. This misleading guidance necessitates documentation updates to align with the correct metric tracking approach compatible with Keras 3.
• [issues/110884]

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: 0

Summarized Issues:

As of our latest update, there were no issues closed in the project this week.

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: 9

Key Open Pull Requests

1. Fix SparseConcat crash on integer overflow in shape computation: This pull request fixes a fatal crash in the SparseConcat operation caused by integer overflow during shape computation by adding explicit overflow validation, replacing unchecked dimension growth with safer methods, preventing fatal aborts, and including a regression test to ensure proper error handling for extremely large input shapes.

• URL: pull/110622

• Associated Commits: 44840, 83803, dcae9, b47ac

2. added same size validation for var and accum_update in ApplyAdadeltaOp. Fixes #104078: This pull request adds a validation check in the ApplyAdadeltaOp kernel to ensure that the input variables var and accum_update are of the same size, addressing a potential Null Pointer Dereference vulnerability on the GPU.

• URL: pull/110559

• Associated Commits: 99dfe

3. Fix resource misclassification for unsorted _read_only_resource_inputs: This pull request fixes a logical flaw in the TensorFlow codebase where the functions _get_read_only_resource_input_indices_op and get_read_write_resource_inputs incorrectly classified read-only resources as read-write when the _read_only_resource_inputs attribute list was unsorted, by replacing the pointer-based linear scan with a more efficient and order-agnostic set-based lookup to prevent spurious control dependencies and ensure correct parallel operation execution.

• URL: pull/110665

• Associated Commits: e60c4

Other Open Pull Requests

• Bug fixes in kernel and operation validation: Multiple pull requests address critical bug fixes in TensorFlow operations and kernels. These include adding checks in the Bincount CUDA kernels to prevent negative bin values and GPU memory access errors, fixing a CHECK failure in MatrixDiagOp by validating tensor dimensions to avoid hard aborts, and correcting resource input categorization by sorting _read_only_resource_inputs to prevent inconsistent classification.
• [pull/110681, pull/110822, pull/110850]

• Build and dependency updates: Some pull requests focus on improving build configurations and dependencies. These updates add the absl::log library to TensorFlow Lite for enhanced logging and update the lit PyPI target reference to support bzlmod integration in the TensorFlow project.
• [pull/110784, pull/110787]

• Code refactoring for safety and compatibility: One pull request refactors the ShapeUtil.cc file to improve memory safety and reliability. It replaces manual array lookups with size checks, enhances overflow detection for large dimensions, refines error handling to prevent crashes during shape creation, and updates layout assignments for better compatibility with modern XLA standards.
• [pull/110833]

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: 2

Key Closed Pull Requests

1. fix: tf.data debug mode breaks dataset.save() due to shard_func returning None: This pull request addresses a bug in TensorFlow's tf.data debug mode where calling dataset.save() fails due to the shard_func returning None, by modifying the dummy shard function to return a valid tensor value (initially np.int64(0), later changed to a plain int) to prevent a TypeError during tracing.

• URL: pull/110735

• Associated Commits: ebb55, 3aba0, bd19a

• Associated Commits: ebb55, 3aba0, bd19a

2. Remove TensorBoard as hard dependency from r2.21: This pull request removes TensorBoard as a mandatory install-time dependency from the TensorFlow r2.21 release, making it an optional package that users must explicitly install if needed, and updates the build and lockfiles accordingly to reflect this change.

• URL: pull/110709

• Associated Commits: 691d9

• Associated Commits: 691d9

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.