* <p><b>Java 9 users:</b> some use cases can be written with calls to {@code optional.stream()}.
   *
   * @since 11.0
   */
  public abstract Set<T> asSet();

  /**
   * If the instance is present, it is transformed with the given {@link Function}; otherwise,
   * {@link Optional#absent} is returned.
   *
   * <p><b>Comparison to {@code java.util.Optional}:</b> this method is similar to Java 8's {@code

import org.jetbrains.kotlin.fir.resolve.providers.toSymbol
import org.jetbrains.kotlin.fir.resolve.scope
import org.jetbrains.kotlin.fir.resolve.toSymbol
import org.jetbrains.kotlin.fir.resolve.transformers.FirImportResolveTransformer
import org.jetbrains.kotlin.fir.scopes.CallableCopyTypeCalculator
import org.jetbrains.kotlin.fir.scopes.impl.FirExplicitSimpleImportingScope

   * fromCollection} supports them.
   *
   * <p>The returned collection isn't threadsafe or serializable, even if {@code fromCollection} is.
   *
   * <p>When a live view is <i>not</i> needed, it may be faster to copy the transformed collection
   * and use the copy.
   *
   * <p>If the input {@code Collection} is known to be a {@code List}, consider {@link
   * Lists#transform}. If only an {@code Iterable} is available, use {@link Iterables#transform}.

                                               void* device_info);

  // Method to execute an operation.
  //
  // Arguments provide enough information to reconstruct the original `TFE_Op`,
  // or construct a transformed version, by inspecting the passed `op`.
  //
  // TFE_OpGetDevice(op) records the original placement of the operation. It may
  // be an empty string if no device was explicitly requested, but will

          <name>pomFile</name>
          <version>4.1.0+</version>
          <required>false</required>
          <description>Originating POM file</description>
          <type>DOM</type> <!-- This is transformed to a File/Path in the template -->
        </field>

        <!-- ====================================================================== -->

*   `Optimizer.__init__` now accepts a `gradient_aggregator` to allow for
    customization of how gradients are aggregated across devices, as well as
    `gradients_transformers` to allow for custom gradient transformations (such
    as gradient clipping).
*   Improvements to Keras preprocessing layers:
    *   TextVectorization can now accept a vocabulary list or file as an init
        arg.