&ned,tpyrctfihs,?ppa-rettalp,s&egap,r&ahc21,ekrow,??tilper:.&arik,d&eer,i,racip,?e&kip,saelererp,?frow,gnigats,k&cops,rik,?labolg,mik,o&do,ksis,?re&hcra,k&c&ah,ut,?ir,??s&enob,irap,maet,?tiprat,ulus,y&awenaj,elsew,ranac,??,vr&esi,uc,?weiverpbuhtig,xdom,ylf,??ih?l!.&di?fnoc?gro?lim?moc?nsa?ten?ude?vog???m!.&eman?gro?lim?m&oc?uesum??o&fni?r&ea?p???pooc?t&en?ni??ude?vog?zib???o&g?m??rt?s!.&bog?der?gro?moc?ude???t!.&arukas,bew-eht-no,morf,naht-&esrow,retteb,?sndnyd,?d?i?won??uqhv--nx??w&a!.moc?hs?l?...

    `Dataset.apply()`.
*   Remove `tf.contrib.data.Iterator.from_dataset()` method. Use
    `Dataset.make_initializable_iterator()` instead.
*   Remove seldom used and unnecessary `tf.contrib.data.Iterator.dispose_op()`.
*   Reorder some TF-GAN loss functions in a non-backwards compatible way.

## Known Issues