// write, we won't see copying because of the new threshold. So it is always OK to use the
      // calculation here.
      return (int) Math.ceil(expectedSize / 0.75);
    }
    return Integer.MAX_VALUE; // any large value
  }

  /**
   * Creates a <i>mutable</i>, empty, insertion-ordered {@code LinkedHashMap} instance.
   *

        checkNotNull(cell.getValue(), "value");
        @SuppressWarnings("unchecked") // all supported methods are covariant
        Cell<R, C, V> immutableCell = (Cell<R, C, V>) cell;
        cells.add(immutableCell);
      } else {
        put(cell.getRowKey(), cell.getColumnKey(), cell.getValue());
      }
      return this;
    }

    /**

    *   tf.optimizers => tf.keras.optimizers
    *   Add tf.keras.layers.AbstractRNNCell as the preferred implementation of
        RNN cell for TF v2. User can use it to implement RNN cell with custom
        behavior.
    *   Adding `clear_losses` API to be able to clear losses at the end of
        forward pass in a custom training loop in eager.

cel.dev/expr v0.15.0 h1:O1jzfJCQBfL5BFoYktaxwIhuttaQPsVWerH9/EEKx0w=
cel.dev/expr v0.15.0/go.mod h1:TRSuuV7DlVCE/uwv5QbAiW/v8l5O8C4eEPHeu7gf7Sg=
cloud.google.com/go v0.26.0/go.mod h1:aQUYkXzVsufM+DwF1aE+0xfcU+56JwCaLick0ClmMTw=
cloud.google.com/go v0.34.0/go.mod h1:aQUYkXzVsufM+DwF1aE+0xfcU+56JwCaLick0ClmMTw=
cloud.google.com/go v0.38.0/go.mod h1:990N+gfupTy94rShfmMCWGDn0LpTmnzTp2qbd1dvSRU=
cloud.google.com/go v0.112.2 h1:ZaGT6LiG7dBzi6zNOvVZwacaXlmf3lRqnC4DQzqyRQw=

* Update example upgrade in docs main page. PR [#795](https://github.com/tiangolo/fastapi/pull/795) by [@cdeil](https://github.com/cdeil).
* Fix callback handling for sub-routers. PR [#792](https://github.com/tiangolo/fastapi/pull/792) by [@jekirl](https://github.com/jekirl).