如何运行新的Jupyter Notebook文件，该文件不属于Docker中的预构建Docker映像? [英] How to run a new Jupyter Notebook file that's not part of a pre-built docker image in docker?

问题描述

我是Docker的新手.为了参加Udacity深度学习课程，我必须使用Docker在Windows计算机上设置TensorFlow.(尽管TensorFlow现在在Windows上可用，但它仅支持Python 3.5，但是Udacity课程材料需要Python 2.7.因此，我必须坚持使用TensorFlow的Docker方法.)

要进行作业，我遵循了

但是，我现在想要的是运行该课程的最终项目，该项目不属于预先构建的Udacity docker映像.我怎样才能做到这一点?

I am new to Docker. In order to take the Udacity Deep Learning course, I had to set up TensorFlow on my Windows machine using Docker. (Although TensorFlow is now available on Windows, it only supports Python 3.5, however the Udacity course material requires Python 2.7. Therefore, I have to stick with the Docker way of using TensorFlow.)

To work on the assignments, I followed the instructions here as detailed below:

• First, I installed docker toolbox.
• Then, I launch Docker using the Docker Quickstart Terminal. For the first time, I ran:

docker run -p 8888:8888 --name tensorflow-udacity -it gcr.io/tensorflow/udacity-assignments:0.6.0.

Each time after, I just run this in my docker terminal:

docker start -ai tensorflow-udacity

• Finally, in the address bar, with http://192.168.99.100:8888 I get the assignment Jupyter notebooks up and running (see image below).

However, what I want now is to run the final project of the course which is not part of the pre-built Udacity docker image. How can I do that? The final project can be found here, with the "digit_recognition.ipynb" specifically being the file to run in docker.

Any guidance is much appreciated.

解决方案

An alternative and much easier way is this:

• Just start up your container like this: $ docker start -ai tensorflow-udacity
• Then, click the upload button and locate the final project iPython Notebook file and upload it.

That's it. Whatever changes you make will be retained and you'll be able to see the new file in the container going forward!

这篇关于如何运行新的Jupyter Notebook文件，该文件不属于Docker中的预构建Docker映像?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！