*   (since v2.7) Run `tf.scatter_nd` and other related scatter functions,
        such as `tf.tensor_scatter_nd_update`, on CPU (with significant
        performance penalty).
    *   Add determinism-unimplemented exception-throwing to the following ops.
        When op-determinism is expected (i.e. after
        `tf.config.experimental.enable_op_determinism` has been called), an

   * probability.
   *
   * <p>The constructed {@code BloomFilter} will be serializable if the provided {@code Funnel<T>}
   * is.
   *
   * <p>It is recommended that the funnel be implemented as a Java enum. This has the benefit of
   * ensuring proper serialization and deserialization, which is important since {@link #equals}
   * also relies on object identity of funnels.
   *

        throw new UnsupportedOperationException("Not implemented yet");
    }

    /**
     * Searches a view asynchronously (not implemented).
     *
     * @param request the search view request
     * @return never returns as this operation is not implemented
     * @throws UnsupportedOperationException always thrown as this operation is not implemented
     */
    @Override

        // Name should be based on actual class, not parent
        assertEquals("extended_test", searcher.getName());
    }

    /**
     * Test that search method must be implemented.
     */
    @Test
    public void test_searchMethodAbstract() {
        final TestSearcher searcher = new TestSearcher();
        // Should be able to call search

/**
 * This class provides a skeletal implementation of {@link ValueGraph}. It is recommended to extend
 * this class rather than implement {@link ValueGraph} directly.
 *
 * <p>The methods implemented in this class should not be overridden unless the subclass admits a
 * more efficient implementation.
 *
 * @author James Sexton
 * @param <N> Node parameter type
 * @param <V> Value parameter type
 * @since 20.0

/**
 * Tests our AtomicHelper fallback strategy in AggregateFutureState.
 *
 * <p>On different platforms AggregateFutureState uses different strategies for its core
 * synchronization primitives. The strategies are all implemented as subtypes of AtomicHelper and
 * the strategy is selected in the static initializer of AggregateFutureState. This is convenient
 * and performant but introduces some testing difficulties. This test exercises the two fallback

  }

  public void testConstraint() {
    /*
     * We'll give ourselves a pass on testing all the possible ways of breaking the constraint,
     * because we know that newClassMap() is implemented using ConstrainedMap which is itself
     * well-tested. A purist would object to this, but what can I say, we're dirty cheaters.
     */
    map.put(Integer.class, new Integer(5));

  static final HashFunction MURMUR3_128 = new Murmur3_128HashFunction(0);

  static final HashFunction GOOD_FAST_HASH_128 = new Murmur3_128HashFunction(GOOD_FAST_HASH_SEED);

  // TODO(user): when the shortcuts are implemented, update BloomFilterStrategies
  private final int seed;

  Murmur3_128HashFunction(int seed) {
    this.seed = seed;
  }

  @Override
  public int bits() {
    return 128;
  }

  @Override

/**
 * Tests our AtomicHelper fallback strategy in AggregateFutureState.
 *
 * <p>On different platforms AggregateFutureState uses different strategies for its core
 * synchronization primitives. The strategies are all implemented as subtypes of AtomicHelper and
 * the strategy is selected in the static initializer of AggregateFutureState. This is convenient
 * and performant but introduces some testing difficulties. This test exercises the two fallback

* Why is this change done? What's the use case?
* For user-facing features, what will the API look like?
* What test cases should it have? What could go wrong?
* How will it roughly be implemented? We'll happily provide code pointers to save you time.

We may ask you to answer these questions directly in the GitHub issue or (for large changes) in a shared Google Doc.