TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Installing TensorFlow 1.4.0 on MacOS with CUDA Support

47 点作者 earcar超过 7 年前

4 条评论

applecrazy超过 7 年前
For me, the number one gripe with TF on Mac is the lack of OpenCL support. With the new iMac Pro coming out with an AMD Vega GPU, and the 2016 MBP having Intel graphics, I feel like it should at least be on their list of things to do. But then again, it's probably not a good idea to train on a Mac.
评论 #15734680 未加载
qixiang109超过 7 年前
I run bazel build and I get this:<p>ERROR: &#x2F;private&#x2F;var&#x2F;tmp&#x2F;_bazel_qixiang&#x2F;5571543a3fba611a0b0ca8efd02f3a37&#x2F;external&#x2F;protobuf_archive&#x2F;BUILD:265:1: C++ compilation of rule &#x27;@protobuf_archive&#x2F;&#x2F;:js_embed&#x27; failed (Exit 1). gcc-7: error: unrecognized command line option &#x27;-fcolor-diagnostics&#x27; gcc-7: error: unrecognized command line option &#x27;-Wthread-safety&#x27;; did you mean &#x27;-fthread-jumps&#x27;? gcc-7: error: unrecognized command line option &#x27;-Wself-assign&#x27;; did you mean &#x27;-Wcast-align&#x27;? Target &#x2F;&#x2F;tensorflow&#x2F;tools&#x2F;pip_package:build_pip_package failed to build<p>any help?
评论 #15741885 未加载
srcmap超过 7 年前
Should also publish the benchmarks of best of class machines running TensorFlow - for Linux, Windows and Mac HWs.<p>Show Apple why they need to release better Mac Desktop&#x2F;Laptop HW or risk loss the heart and mind of next generation of developers.
nl超过 7 年前
It&#x27;s worth noting that <i>any</i> TF user (including those on NVidia Macs) can get significant speedups (20%+) by recompiling to use all the features of your CPU.<p>TF even warns you about this when you start up.
评论 #15734701 未加载