?m??yjnoh??u&kobmes?oppah????in,?o!.&a&dakatognub?m&asah?ihsemih??su?t&ekat?i&h?o????e&onokok?ustimak??i&jih?k&asinuk?ias?usu??mukust??onoognub?u&fuy?juk?ppeb?suk?????nayn,?wa&ga&k!.&a&mihsoan?rihotok?waga&kihsagih?ya???emaguram?i&j&nonak?ustnez??kunas?monihcu??o&hsonot?nnam?yotim??u&st&amakat?odat??zatu????nak!.&a&dustam?kus&okoy?tarih??maz?nibe?r&a&gihsaimanim?h&esi?imagas??wa&do?guy???u&im?kamak???tikamay?wa&k&ia?oyik?umas??sijuf??yimonin??e&nokah?saya??i&akan?esiak?gusta?hsuz?kasagihc?o?ukus...

*   `tf.experimental.StructuredTensor`

    *   Introduced `tf.experimental.StructuredTensor`, which provides a flexible
        and Tensorflow-native way to encode structured data such as protocol
        buffers or pandas dataframes.

*   `tf.keras`:

    *   Added method `get_metrics_result()` to `tf.keras.models.Model`.
        *   Returns the current metrics values of the model as a dict.