kale
kale copied to clipboard
kubeflow-kale
ModuleNotFoundError: No module named 'kale.rpc'; 'kale' is not a package
Setup: MacOS 10.15.6 JupyterLab: 1.02, 1.1.3, 1.1.4 Python: 3.7.7 Node.js: 10.13.0, npm: 6.4.1 Kubeflow-Kale-labextension installed
This is my first time to try Kale and have googled for solutions with no progress. The following error kept poping up with a button "Refresh" when opening a notebook in the lab. Is this a bug or am I missing anything? Thanks!
A Kernel Error has occurred Browser: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_6) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/84.0.4147.135 Safari/537.36
Type: Kernel
Method: log.setup_logging()
Message: ModuleNotFoundError: No module named 'kale.rpc'; 'kale' is not a package
Details: [ "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", "\u001b[0;31mModuleNotFoundError\u001b[0m Traceback (most recent call last)", "\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0;32mfrom\u001b[0m \u001b[0mkale\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrpc\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrun\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mrun\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0m__kale_rpc_run\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 2\u001b[0m \u001b[0m__kale_rpc_result\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m__kale_rpc_run\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m"log.setup_logging"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'e30='\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'e30='\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;31mModuleNotFoundError\u001b[0m: No module named 'kale.rpc'; 'kale' is not a package" ]
Hi @scottieliu,
It looks like the Kale Python package is not installed. Take into consideration that the extension cannot work without the backend part of Kale.
How did you install it? Can you verify you have installed both parts (lab extension and python package)?
I have a local project using Jupyter notebook and it works fine. My goal was to package it with Kale and JupyterLab. I did the following:
pip install jupyterlab==1.2.6 # tried other 1.* versions, too. pip install kubeflow-kale jupyter labextension install kubeflow-kale-labextension pip3 install kubeflow-kale kfp kfp-server-api urllib3
What is the second part of python package (is it Python)?
I was able to enable the Kale Deployment Panel and run "Compile and Run/Compile and Save/Compile Upload" with green check marks on the tasks in the panel. But I did not see any new UI changes on the project related to the pipeline.
I see the following errors in the shell window:
[E 21:58:10.184 LabApp] Notebook JSON is invalid: 'execution_count' is a required property
Failed validating 'required' in execute_result:
On instance['cells'][74]['outputs'][0]:
{'data': {'text/html': '<div>\n'
'<style scoped>\n'
' .dataframe tbody tr th:only-of-type {\n'
' ...'},
'metadata': {},
'output_type': 'execute_result'}
The installation seems good. I suggest you either use pip or pip3, mixing the two (even if they point to the same python installation) may lead to frustration and bugs that are hard to be traced.
But I did not see any new UI changes on the project related to the pipeline
Could you explain this a bit? What new UI changes were you expecting to see?
Also note that with the commands you provided you are installing the latest releases, so changes on master branch merged after the release won't be there.
I see the following errors in the shell window: [E 21:58:10.184 LabApp] Notebook JSON is invalid: 'execution_count' is a required property
Could you provide some more information on how you constructed the notebook? What tools did you use? Did you by any chance manually edit it? Kale most probably doesn't touch this property (it does when you recover a pipeline step's state from a Rok snapshot, but it doesn't remove it making the notebook JSON invalid).
I'm having the same issue; including the lack of UI changes.
I was expecting to find a "Kale" icon in the left-hand menu. Different online tutorials indicate that should be present and its from there that you initiate the pipeline conversion.
I used the following for my Dockerfile to run the instance:
FROM gcr.io/kubeflow-images-public/tensorflow-1.15.2-notebook-gpu:1.0.0
USER root
RUN pip3 install --upgrade pip
RUN pip3 install kubeflow-kale
RUN jupyter labextension install kubeflow-kale-launcher
RUN jupyter labextension list
RUN chown jovyan -R /home/jovyan
USER 1000
I found some other install instructions and modified my Dockerfile to the following and it now works:
FROM gcr.io/kubeflow-images-public/tensorflow-1.15.2-notebook-gpu:1.0.0
USER root
RUN cd /tmp && git clone https://github.com/kubeflow-kale/kale
RUN cd /tmp/kale/backend && python3 ./setup.py install
RUN pip3 install --upgrade pip
RUN pip3 install kubeflow-kale
RUN jupyter labextension install kubeflow-kale-launcher
RUN jupyter labextension list
RUN chown jovyan -R /home/jovyan
USER 1000
@LRSutton according to https://github.com/kubeflow-kale/kale/issues/187#issuecomment-688742762 kubeflow-kale-launcher is deprecated and you should use kubeflow-kale-labextension
