return ensureInUnitRange(sumOfProductsOfDeltas / Math.sqrt(productOfSumsOfSquaresOfDeltas));
  }

  /**
   * Returns a linear transformation giving the best fit to the data according to <a
   * href="http://mathworld.wolfram.com/LeastSquaresFitting.html">Ordinary Least Squares linear
   * regression</a> of {@code y} as a function of {@code x}. The count must be greater than one, and

    }
    return this;
  }

  /**
   * Adds each member of {@code elements} as a candidate for the top {@code k} elements. This
   * operation takes amortized linear time in the length of {@code elements}.
   *
   * <p>If all input data to this {@code TopKSelector} is in a single {@code Iterable}, prefer

  // Returns the non-null array of replacements for fast lookup.
  char[][] getReplacementArray() {
    return replacementArray;
  }

  // Creates a replacement array from the given map. The returned array is a
  // linear lookup table of replacement character sequences indexed by the
  // original character value.
  @VisibleForTesting
  static char[][] createReplacementArray(Map<Character, String> map) {

    return ensureInUnitRange(sumOfProductsOfDeltas / Math.sqrt(productOfSumsOfSquaresOfDeltas));
  }

  /**
   * Returns a linear transformation giving the best fit to the data according to <a
   * href="http://mathworld.wolfram.com/LeastSquaresFitting.html">Ordinary Least Squares linear
   * regression</a> of {@code y} as a function of {@code x}. The count must be greater than one, and

    return ensureInUnitRange(sumOfProductsOfDeltas / Math.sqrt(productOfSumsOfSquaresOfDeltas));
  }

  /**
   * Returns a linear transformation giving the best fit to the data according to <a
   * href="http://mathworld.wolfram.com/LeastSquaresFitting.html">Ordinary Least Squares linear
   * regression</a> of {@code y} as a function of {@code x}. The count must be greater than one, and

    return ensureInUnitRange(sumOfProductsOfDeltas / Math.sqrt(productOfSumsOfSquaresOfDeltas));
  }

  /**
   * Returns a linear transformation giving the best fit to the data according to <a
   * href="http://mathworld.wolfram.com/LeastSquaresFitting.html">Ordinary Least Squares linear
   * regression</a> of {@code y} as a function of {@code x}. The count must be greater than one, and

background-color: inherit !important;\n}\n\n[data-widget=\"expandable-table\"] {\n  cursor: pointer;\n}\n\n[data-widget=\"expandable-table\"] i.expandable-table-caret {\n  transition: -webkit-transform 0.3s linear;\n  transition: transform 0.3s linear;\n  transition: transform 0.3s linear, -webkit-transform 0.3s linear;\n}\n\n[data-widget=\"expandable-table\"][aria-expanded=\"true\"] i.expandable-table-caret[class*=\"right\"] {\n  -webkit-transform: rotate(90deg);\n  transform: rotate(90deg);\n}\n\n[data-widg...

    checkNotNull(presentBehavior);
    checkNotNull(absentBehavior);
    if (!(list instanceof RandomAccess)) {
      list = new ArrayList<>(list);
    }
    // TODO(lowasser): benchmark when it's best to do a linear search

    int lower = 0;
    int upper = list.size() - 1;

    while (lower <= upper) {
      int middle = (lower + upper) >>> 1;
      int c = comparator.compare(key, list.get(middle));

    }
  }

  private static boolean equal(@Nullable Object a, @Nullable Object b) {
    return a == b || (a != null && a.equals(b));
  }

  // This one-liner saves us from some ugly casts
  protected Entry<K, V> entry(K key, V value) {
    return mapEntry(key, value);
  }

  @Override
  protected void expectContents(Collection<Entry<K, V>> expected) {

    }
  }

  private static boolean equal(@Nullable Object a, @Nullable Object b) {
    return a == b || (a != null && a.equals(b));
  }

  // This one-liner saves us from some ugly casts
  protected Entry<K, V> entry(K key, V value) {
    return mapEntry(key, value);
  }

  @Override
  protected void expectContents(Collection<Entry<K, V>> expected) {