设为首页收藏本站

LUPA开源社区

 找回密码
 注册
文章 帖子 博客
LUPA开源社区 首页 业界资讯 软件追踪 查看内容

TensorFlow 1.8.0-rc0发布,bug修复版本

2018-4-14 20:59| 发布者: joejoe0332| 查看: 564| 评论: 0|原作者: oschina|来自: oschina

摘要: TensorFlow 1.8.0-rc0 发布,此版本包括很多性能改进和bug修复。主要特性和改进包括:主要特性和改进Can now passtf.contrib.distribute.MirroredStrategy()totf.estimator.RunConfig()to run an Estimator model on ...

TensorFlow 1.8.0-rc0 发布,此版本包括很多性能改进和bug修复。主要特性和改进包括:

主要特性和改进

  • Can now pass tf.contrib.distribute.MirroredStrategy() to tf.estimator.RunConfig() to run an Estimator model on multiple GPUs on one machine.

  • Add tf.contrib.data.prefetch_to_device(), which supports prefetching to GPU memory.

  • Added Gradient Boosted Trees as pre-made Estimators: BoostedTreesClassifier, BoostedTreesRegressor.

  • Add 3rd generation pipeline config for Cloud TPUs which improves performance and usability.

  • tf.contrib.bayesflow is moving out to it's own repo.

  • Added tf.contrib.{proto,rpc} to allow generic proto parsing and RPC communication.

Bug 修复和其他改变

  • tf.data:

    • Add tf.contrib.data.prefetch_to_device, which enables prefetching dataset elements to GPU memory.

    • Add tf.contrib.data.AUTOTUNE, which allows the tf.data runtime to automatically tune the prefetch buffer sizes based on your system and environment.

    • Add tf.contrib.data.make_csv_dataset for building datasets of CSV files.

  • Eager Execution:

    • With eager execution Datasets can now be used as standard python iterators (for batch in dataset:). Both Dataset.__iter__() and Dataset.make_one_shot_iterator() can now be used to create iterators when eager execution is enabled.

    • Automatic device placement has been enabled (i.e., use a GPU if available automatically, without requiring an explicit with tf.device(“/gpu:0”)) (Fixes #14133)

    • tf.GradientTape has moved out of contrib.

完整内容请查看发布主页

下载地址:


酷毙

雷人

鲜花

鸡蛋

漂亮
  • 快毕业了,没工作经验,
    找份工作好难啊?
    赶紧去人才芯片公司磨练吧!!

最新评论

关于LUPA|人才芯片工程|人才招聘|LUPA认证|LUPA教育|LUPA开源社区 ( 浙B2-20090187 浙公网安备 33010602006705号   

返回顶部