}

  tensorflow::shape_inference::ShapeHandle shape = ic->output(output.index);

  int rank = -1;
  if (ic->RankKnown(shape)) {
    rank = ic->Rank(shape);
  }

  if (num_dims != rank) {
    status->status = InvalidArgument("Expected rank is ", num_dims,
                                     " but actual rank is ", rank);
    return;
  }

  if (num_dims == 0) {
    // Output shape is a scalar.

        final ArtifactStatus that = (ArtifactStatus) o;

        return rank == that.rank;
    }

    @Override
    public int hashCode() {
        return rank;
    }

    @Override
    public String toString() {
        return key;
    }

    @Override
    public int compareTo(ArtifactStatus s) {
        return rank - s.rank;
    }

    this.rankMap = rankMap;
  }

  @Override
  public int compare(T left, T right) {
    return rank(left) - rank(right); // safe because both are nonnegative
  }

  private int rank(T value) {
    Integer rank = rankMap.get(value);
    if (rank == null) {
      throw new IncomparableValueException(value);
    }
    return rank;
  }

  @Override
  public boolean equals(@Nullable Object object) {

    }

    /** Equivalent to {@code newReentrantReadWriteLock(rank, false)}. */
    public ReentrantReadWriteLock newReentrantReadWriteLock(E rank) {
      return newReentrantReadWriteLock(rank, false);
    }

    /**
     * Creates a {@link ReentrantReadWriteLock} with the given fairness policy and rank. The values

    }

    /** Equivalent to {@code newReentrantReadWriteLock(rank, false)}. */
    public ReentrantReadWriteLock newReentrantReadWriteLock(E rank) {
      return newReentrantReadWriteLock(rank, false);
    }

    /**
     * Creates a {@link ReentrantReadWriteLock} with the given fairness policy and rank. The values

  const auto& shape = reader->GetVariableToShapeMap().at(name);
  int rank = shape.dims();
  if (num_dims != rank) {
    status->status = InvalidArgument("Expected rank is ", num_dims,
                                     " but actual rank is ", rank);
    return;
  }
  for (int i = 0; i < num_dims; i++) {
    dims[i] = shape.dim_size(i);
  }
}

    final T value;
    final int rank;

    Composite(T value, int rank) {
      this.value = value;
      this.rank = rank;
    }

    // natural order is by rank only; the test will compound() this with the
    // order of 't'.
    @Override
    public int compareTo(Composite<T> that) {
      return Integer.compare(rank, that.rank);
    }

    final T value;
    final int rank;

    Composite(T value, int rank) {
      this.value = value;
      this.rank = rank;
    }

    // natural order is by rank only; the test will compound() this with the
    // order of 't'.
    @Override
    public int compareTo(Composite<T> that) {
      return Integer.compare(rank, that.rank);
    }

        as `tf.tpu.experimental.embedding.serving_embedding_lookup` which can
        take arbitrary rank of dense and sparse tensor. For ragged tensor,
        though the input tensor remains to be rank 2, the activations now can be
        rank 2 or above by specifying the output shape in the feature config or
        via the build method.

*   Add

But still, the final experience is not the best. 😞

---

This would be the parallel equivalent story for burgers. 🍔

For a more "real life" example of this, imagine a bank.

Up to recently, most of the banks had multiple cashiers 👨‍💼👨‍💼👨‍💼👨‍💼 and a big line 🕙🕙🕙🕙🕙🕙🕙🕙.

All of the cashiers doing all the work with one client after the other 👨‍💼⏯.