TF_LOCKS_EXCLUDED(session->graph->mu, session->mu);

std::string getTF_OutputDebugString(TF_Output node);

// Set whether to propagate assigned device information when constructing a new
// Graph from a GraphDef. By default assigned device information is not copied
// and is re-computed by the runtime.
inline void TF_ImportGraphDefOptionsSetPropagateDeviceSpec(

 * limitations under the License.
 * =============================================================================
 */

/** Extracts PR from commit message and creates a GitHub Issue on Rollback of PR
  Created issue is assigned to original PR owner and reviewer.

  @param {!object}
    github enables querying for PR and also create issue using rest endpoint
    context has the commit message details in the payload

        uses: actions/stale@6f05e4244c9a0b2ed3401882b05d701dd0a7289b # v7.0.0
        with:
          #Comma separated list of labels that can be assigned to issues to exclude them from being marked as stale
          exempt-issue-labels: 'override-stale'
          #Comma separated list of labels that can be assigned to PRs to exclude them from being marked as stale
          exempt-pr-labels: "override-stale"

  return domain;
};

/** For trusted parters like Intel, we want to auto-run tests
    This allows us to reduce the delay to external partners
    Add Labels - kokoro:force-run
    The PR is also assigned to specific teams to fast track review
    Additional reviewers can be added manually based on PR contents
  @param {!object}
    github enables querying for PR and also create issue using rest endpoint

If you have improvements to TensorFlow, send us your pull requests! For those
just getting started, GitHub has a
[how-to](https://help.github.com/articles/using-pull-requests/).

TensorFlow team members will be assigned to review your pull requests. Once the
pull requests are approved and pass continuous integration checks, a TensorFlow
team member will apply `ready to pull` label to your change. This means we are

// a map (`tensorflow::eager::TensorTape`) from the wrapped tensor to the id of
// the op that produced it (or -1 if this tensor was watched using
// `GradientTape::Watch`.) The op_id is simply a unique index assigned to each
// op executed under the tape. A separate map (`tensorflow::eager::OpTape`)
// maintains the map from `op_id` to a `OpTapeEntry` which stores the `op_type`,

        values every time they are called (for a given variable shape). For
        unseeded initializers (`seed=None`), a random seed will be created and
        assigned at initializer creation (different initializer instances get
        different seeds).
    *   An unseeded initializer will raise a warning if it is reused (called)