* is not guaranteed to return zero when the dataset consists of the same pair of values multiple
   * times, due to numerical errors.
   *
   * <h3>Non-finite values</h3>
   *
   * <p>If the dataset contains any non-finite values ({@link Double#POSITIVE_INFINITY}, {@link
   * Double#NEGATIVE_INFINITY}, or {@link Double#NaN}) then the result is {@link Double#NaN}.
   *

 *
 * <h3>Other types of preconditions</h3>
 *
 * <p>Not every type of precondition failure is supported by these methods. Continue to throw
 * standard JDK exceptions such as {@link java.util.NoSuchElementException} or {@link
 * UnsupportedOperationException} in the situations they are intended for.
 *
 * <h3>Non-preconditions</h3>
 *

import java.util.Collection;
import java.util.LinkedHashMap;
import java.util.Map;

/**
 * Provides a fluent API for calculating <a
 * href="http://en.wikipedia.org/wiki/Quantile">quantiles</a>.
 *
 * <h3>Examples</h3>
 *
 * <p>To compute the median:
 *
 * <pre>{@code
 * double myMedian = median().compute(myDataset);
 * }</pre>
 *
 * where {@link #median()} has been statically imported.
 *

 *
 * <p>This interface is meant to be used as the type of a parameter to graph algorithms (such as
 * topological sort) that only need a way of accessing the predecessors of a node in a graph.
 *
 * <h3>Usage</h3>
 *
 * Given an algorithm, for example:
 *
 * <pre>{@code
 * public <N> someGraphAlgorithm(N startNode, PredecessorsFunction<N> predecessorsFunction);
 * }</pre>
 *

 *
 * <p>This interface is meant to be used as the type of a parameter to graph algorithms (such as
 * breadth first traversal) that only need a way of accessing the successors of a node in a graph.
 *
 * <h3>Usage</h3>
 *
 * Given an algorithm, for example:
 *
 * <pre>{@code
 * public <N> someGraphAlgorithm(N startNode, SuccessorsFunction<N> successorsFunction);
 * }</pre>
 *

 *         .transform(User::getId, directExecutor())
 *         .transform(ActivityService::isLoggedIn, threadPool)
 *         .catching(RpcException.class, e -> false, directExecutor());
 * }</pre>
 *
 * <h3>Alternatives</h3>
 *
 * <h4>Frameworks</h4>
 *
 * <p>When chaining together a graph of asynchronous operations, you will often find it easier to
 * use a framework. Frameworks automate the process, often adding features like monitoring,

 * href="https://github.com/google/guava/wiki/GraphsExplained#choosing-the-right-graph-type">
 * "Choosing the right graph type"</a> section of the Guava User Guide for more details.
 *
 * <h3>Capabilities</h3>
 *
 * <p>{@code ValueGraph} supports the following use cases (<a
 * href="https://github.com/google/guava/wiki/GraphsExplained#definitions">definitions of
 * terms</a>):
 *
 * <ul>
 *   <li>directed graphs

 *
 * <p>Like any other {@code Map<Class, Object>}, this map may contain entries for primitive types,
 * and a primitive type and its corresponding wrapper type may map to different values.
 *
 * <h3>Implementations</h3>
 *
 * <ul>
 *   <li>{@link ImmutableClassToInstanceMap}
 *   <li>{@link MutableClassToInstanceMap}
 * </ul>
 *
 * <p>To map a generic type to an instance of that type, use {@link

 *
 * <p>The following examples show a simple threadsafe holder expressed using {@code synchronized},
 * {@link ReentrantLock}, and {@code Monitor}.
 *
 * <h3>{@code synchronized}</h3>
 *
 * <p>This version is the fewest lines of code, largely because the synchronization mechanism used
 * is built into the language and runtime. But the programmer has to remember to avoid a couple of

 *
 * // In case bug 12345 happens again we'd rather just die
 * Verify.verify(bill.status() == Status.UNPAID,
 *     "Unexpected bill status: %s", bill.status());
 * }</pre>
 *
 * <h3>Comparison to alternatives</h3>
 *
 * <p><b>Note:</b> In some cases the differences explained below can be subtle. When it's unclear
 * which approach to use, <b>don't worry</b> too much about it; just pick something that seems