ImportError: Missing required package "azureml-dataset-runtime"

使用 notebook 的 conda 设置以及所示的 Dataset.get_by_name/to_pandas_dataframe 单元格作为复现路径。确认每个 %pip install 由哪个 kernel 和 interpreter 接收，然后记录一个可复现的依赖项修复方案；当该单元格能够在没有缺少软件包错误的情况下运行时即完成。

I am having a weird issue in Azure ML.

I like to work in isolated conda environments with its own kernel, so for each little project I create my own environment and kernel like this:

conda create — name myenv
conda activate myenv
conda install pip
conda install ipykernel
python -m ipykernel install --user --name myenv  --display-name myenv

This creates the environment and the kernel, and then it's totally empty which means I have to install all modules myself. (including Azure SDK)

So I tried to install the following:

%pip install azureml
%pip install azureml-core
%pip install pandas
%pip install azureml-dataset-runtime --upgrade

And then I tried to run a simple Load Dataset code

# azureml-core of version 1.0.72 or higher is required
# azureml-dataprep[pandas] of version 1.1.34 or higher is required
from azureml.core import Workspace, Dataset

subscription_id = 'xx'
resource_group = 'mlplayground'
workspace_name = 'mlplayground'

workspace = Workspace(subscription_id, resource_group, workspace_name)

dataset = Dataset.get_by_name(workspace, name='BelgiumRealEstate')
df=dataset.to_pandas_dataframe()

But then I get this error:

Missing required package "azureml-dataset-runtime", which can be installed by running: "/anaconda/envs/LightGBM/bin/python" -m pip install azureml-dataset-runtime --upgrade

And Yes, I tried to run this command, but got errors when trying to install it:

Can someone please enlighten me? Am I doing something wrong?

Thanks a lot.