| 
 TensorfFlow 0.6.0 发布,支持python 3.3  
2015年12月10日 
 
TensorfFlow 0.6.0 发布,更新如下: 
主要特性和提升- 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.
 
  更多内容请看: 
https://github.com/tensorflow/tensorflow/blob/master/RELEASE.md 
 
TensorFlow 是谷歌的第二代机器学习系统,按照谷歌所说,在某些基准测试中,TensorFlow的表现比第一代的DistBelief快了2倍。 
TensorFlow 内建深度学习的扩展支持,任何能够用计算流图形来表达的计算,都可以使用TensorFlow。任何基于梯度的机器学习算法都能够受益于TensorFlow的自动分化(auto-differentiation)。通过灵活的Python接口,要在TensorFlow中表达想法也会很容易。 
TensorFlow 对于实际的产品也是很有意义的。将思路从桌面GPU训练无缝搬迁到手机中运行。 
 
 
 
 
 |