start by opening a new issue describing the bug or feature you're intending to
fix. Even if you think it's relatively minor, it's helpful to know what people
are working on. And as mentioned above, API changes should be discussed
thoroughly before moving to code.

Some examples of types of pull requests that are immediately helpful:

  - Fixing a bug without changing a public API.
  - Fixing or improving documentation.

   *
   * @param url the URL to read from
   * @param charset the charset used to decode the input stream; see {@link StandardCharsets} for
   *     helpful predefined constants
   * @return a string containing all the characters from the URL
   * @throws IOException if an I/O error occurs.
   */
  public static String toString(URL url, Charset charset) throws IOException {

the problem. If possible, please share a link to or attach code demonstrating
the problem.

**Any other info / logs**
Include any logs or source code that would be helpful to diagnose the problem.
If including tracebacks, please include the full traceback. Large logs and files

    steps:
    - name: Checkout code
      uses: actions/checkout@08c6903cd8c0fde910a37f88322edcfb5dd907a8 # v5.0.0
      with:
        ref: ${{ github.event.inputs.release_branch }}
    - name: Get some helpful info for formatting
      id: cherrypick
      run: |
          git config --global user.name "TensorFlow Release Automation"
          git config --global user.email "******@****.***"
          git fetch origin master

the problem. If possible, please share a link to Colab/Jupyter/any notebook.

Also, please include a link to a GraphDef or the model if possible.

**Any other info / logs**

Include any logs or source code that would be helpful to diagnose the problem.
If including tracebacks, please include the full traceback. Large logs and files

   problems later. Be thoughtful and considerate when seeking out the appropriate forum for your
   questions. Those who are asked should be responsive and helpful.

 * **Step down considerately**: Members of every project come and go. When somebody leaves or
   disengages from the project, they should make it known and take the proper steps to ensure that

TFCI_WHL_SIZE_LIMIT=450M
TFCI_WHL_SIZE_LIMIT_ENABLE=1
TFCI_WHL_IMPORT_TEST_ENABLE=1
TFCI_PYTHON_VERIFY_PIP_INSTALL_ARGS=""

# Used to simulate a T:\ drive within the container, to a limited extent,
# via a symlink.
# Helpful since the internal CI utilizes a T:\ drive, part of which is mounted
# to the container, and would result in C:\<path> != T:\<path> mismatches,
# when using variables like `TFCI_OUTPUT_DIR` in `docker exec commands,

        programmers everywhere. If you can give enough detail such that any of us can imagine coming
        across a similar need in our own work, that's extremely helpful in studying how broadly
        useful the proposed change will be.

  - type: textarea
    attributes:
      label: Concrete Use Cases
      description: Please provide use cases that actually came up in the real world.

### 4. (optional) RNN conversion support
If converting TF RNN to TFLite fused RNN ops, please prefix [RNN] in the title.

### 5. (optional) Any other info / logs

limits the impact of these use-after-free bugs by preventing reuse of freed
memory regions until the garbage collector is able to determine that it is
safe. Typically, a use-after-free bug will result in a fault and a helpful
error message, but this package reserves the right to not force a fault on
freed memory. That means a valid implementation of this package is to just
allocate all memory the way the runtime normally would, and in fact, it