课程实例-智能计算系统
目录
踩坑
通用
python 运行报错SyntaxError: Non-ASCII character
SyntaxError: Non-ASCII character '\xe6' in file xxx.py on line xxx, but no encoding declared; see http://python.org/dev/peps/pep-0263/ for details
在文件头上加上# encoding: utf-8
即可
exp4
exp4.4 bazel 报错corrupt installation
错误信息如:
$ bazel version
FATAL: corrupt installation: file '/<workspace>/.cache/bazel/_bazel_root/install/<hash>/A-server.jar' is missing or modified. Please remove '/<workspace>/.cache/bazel/_bazel_root/install/<hash>' and try again.
简单的解决方法是删除缓存文件后重新运行
$ sudo rm -rf /<workspace>/.cache/bazel/_bazel_root/install/<hash>
$ bazel version
需要注意将<workspace>
和<hash>
换成自己的目录
或者可以重新安装 bazel,参考官方文档
$ wget https://releases.bazel.build/0.24.0/rc9/bazel-0.24.0rc9-installer-linux-x86_64.sh
...
$ chmod +x bazel-0.24.0rc9-installer-linux-x86_64.sh
$ ./bazel-0.24.0rc9-installer-linux-x86_64.sh
...
$ export PATH=/usr/local/bin:$PATH
$ bazel version
Build label: 0.24.0rc9
Build target: bazel-out/k8-opt/bin/src/main/java/com/google/devtools/build/lib/bazel/BazelServer_deploy.jar
Build time: Thu Mar 21 19:49:51 2019 (1553197791)
Build timestamp: 1553197791
Build timestamp as int: 1553197791
需要注意安装的版本应在0.24.0
到0.25
,过旧或过新的版本无法编译课程使用的 tensorflow 版本
exp4.4 tensorflow 编译时报错server terminated abruptly
修改编译脚本build_tensorflow-v1.10_mlu.sh
,将jobs_num
从 32 改为 16 即可(更小理论可行,但编译速度会慢;更大未经测试)
$ vi build_tensorflow-v1.10_mlu.sh
...
jobs_num=32 // 改为 jobs_num=16
...