在 windows 中为 jupyter notebook 构建一个 docker 镜像答案

【问题标题】：build a docker image for jupyter notebook in windows在 windows 中为 jupyter notebook 构建一个 docker 镜像
【发布时间】：2019-05-03 04:43:52
【问题描述】：

我想在 windows 中构建一个 docker 镜像，我的 docker 文件是：

FROM jupyter/scipy-notebook

MAINTAINER Jon Krohn <jon@untapt.com>

USER $NB_USER

# install TensorFlow
RUN conda install --quiet --yes 'tensorflow=1.0*'

# install tflearn and keras: 
RUN pip install tflearn==0.3.2
RUN pip install keras==2.0.8

# install NLP packages:
RUN pip install nltk==3.2.4
RUN pip install gensim==2.3.0

# install Reinforcement Learning packages:
RUN pip install gym==0.9.4

但是当我用这个命令构建它时

docker build -t tensorflow-ll-stack .

我收到此错误消息

The environment is inconsistent, please check the package plan carefully
The following packages are causing the inconsistency:

  - conda-forge/noarch::seaborn==0.9.0=py_0
  - conda-forge/linux-64::matplotlib==3.0.3=py37_1
failed

SpecsConfigurationConflictError: Requested specs conflict with configured specs.
  requested specs:
    - tensorflow=1.0
  pinned specs:
    - python=3.7
Use 'conda config --show-sources' to look for 'pinned_specs' and 'track_features'
configuration parameters.  Pinned specs may also be defined in the file
/opt/conda/conda-meta/pinned.


The command '/bin/sh -c conda install --quiet --yes 'tensorflow=1.0*'' returned a non-zero code: 1

感谢您的帮助。

【问题讨论】：

标签： windows docker jupyter-notebook dockerfile

【解决方案1】：

我在 Windows 上设置 docker 映像时遇到了一些类似的问题。对我来说，这是一个 SSL 证书错误，并且在安装之前执行了以下代码。

conda config --set ssl_verify no

所以对于你的情况，你可以试试这个：

FROM jupyter/scipy-notebook

MAINTAINER Jon Krohn <jon@untapt.com>

USER $NB_USER

# install TensorFlow
RUN conda config --set ssl_verify no
RUN conda install --quiet --yes 'tensorflow=1.0*'

# install tflearn and keras: 
RUN pip install tflearn==0.3.2
RUN pip install keras==2.0.8

# install NLP packages:
RUN pip install nltk==3.2.4
RUN pip install gensim==2.3.0

# install Reinforcement Learning packages:
RUN pip install gym==0.9.4

【讨论】：