Feature/jupyter docker #1019
Feature/jupyter docker #1019
Conversation
| @@ -43,5 +43,7 @@ cp -rv /woboq/data $WOBOQ_OUT/../data | |||
| -o $WOBOQ_OUT \ | |||
| -p paddle:/paddle | |||
| /woboq/indexgenerator/codebrowser_indexgenerator $WOBOQ_OUT | |||
|
|
|||
| cd /woboq | |||
| make clean | |||
There was a problem hiding this comment.
make clean will shrink docker image size.
There was a problem hiding this comment.
I want to build a docker env to test this commit, this may take some time, will this block you?
| @@ -0,0 +1,27 @@ | |||
| # from https://github.com/tensorflow/tensorflow/blob/master/tensorflow/tools/docker/jupyter_notebook_config.py | |||
There was a problem hiding this comment.
this file comes from tensorflow
|
|
||
| RUN mkdir -p /opt/run | ||
| COPY ./paddle/scripts/docker/jupyter_notebook_config.py /root/.jupyter/ | ||
| COPY ./paddle/scripts/docker/00_sshd /opt/run/ |
There was a problem hiding this comment.
这里命名成00，01的脚本，貌似是为了按照名字alphabetical order来执行。但是下面 run_all脚本的写法不能保证每次都按照同样的顺序来执行。如果要保证按照字母序，可以参考 http://stackoverflow.com/questions/7992689/bash-how-to-loop-all-files-in-sorted-order
不过我觉得至少现在不需要这么弄把，可以把 00_sshd 和 01_jupyter 写进 run_all 里，顺便在调用一下 jypyter_notebook_config.py 就好了把。
| from IPython.lib import passwd | ||
|
|
||
| c.NotebookApp.ip = '*' | ||
| c.NotebookApp.port = int(os.getenv('PORT', 8888)) |
There was a problem hiding this comment.
这里默认port改改，就不需要标注从Tensorflow来了吧。这个文件的内容都是基于ipython lib的，ipython和jupyter用户都会写，不见得非要受tensorflow启发？
There was a problem hiding this comment.
Done. Also use git rebase purge this config file in git history.
I'm building an image base on this commit and test it.
reyoung
force-pushed
the
feature/jupyter_docker
branch
from
3e16601 to
d9dbfe4
Compare
reyoung
force-pushed
the
feature/jupyter_docker
branch
from
d9dbfe4 to
68c89bc
Compare
|
一个建议： docker/run_all ==> docker/entrypoint 另外，我没有验证docker buld，请 @jacquesqiao 验证再merge吧。谢谢啦！ |
有道理，done |
Add jupyter notebook to docker image.