Python 3.3+ support via changes to python codebase and abilityto specify python version via ./configure.
Some improvements to GPU performance and memory usage:convnet benchmarksroughly equivalent with native cudnn v2 performance. Improvements mostly dueto moving to 32-bit indices, faster shuffling kernels. More improvements tocome in later releases.
Bug 修复
Lots of fixes to documentation and tutorials, many contributedby the public.
271 closed issues on github issues.
向后兼容变化
tf.nn.fixed_unigram_candidate_sampler changed its default 'distortion'attribute from 0.0 to 1.0. This was a bug in the original releasethat is now fixed.