I am receiving a Path Error using Orca in Google Colab to save an image in a Google Colab .ipynb Python Machine Learning model. A Flask App (written in VS Code IDE) uses the Colab trained model to output results to a URL. To be clear, the errors occur in my Colab Python script. This error is very similar to issue #252, however, I am unable to recreate the solution used there.
I Colab, I have created the following recommended import lines, as described in many posts:
!pip install plotly>=4.7.1 !wget https://github.com/plotly/orca/releases/download/v1.2.1/orca-1.2.1-x86_64.AppImage -O /usr/local/bin/orca !chmod +x /usr/local/bin/orca
and
import plotly.graph_objects as go
I then use the following image write command:
fig.write_image('base_pic.svg')
I expect to see 'base_pic.svg' appear as a file when the code runs, but this is not happening.
My error seems to related to a path setting, however, I am unable to find detailed instruction to update and correct this path. Unfortunately, I am quite novice with troubleshooting paths.
The Error:
ValueError Traceback (most recent call last)
in ()
1 #TEST FUNCTION
----> 2 make_picture('AgesAndHeights.pkl',model, floats_string_to_np_array('1, 3, 3.2, '),'x.svg')
5 frames
/usr/local/lib/python3.7/dist-packages/plotly/io/_orca.py in validate_executable()
1109 Searched for the executable 'xvfb-run' on the following path:
1110 {formatted_path}""".format(
-> 1111 formatted_path=formatted_path
1112 )
1113 )
ValueError:
The plotly.io.orca.config.use_xvfb property is set to True, but the
xvfb-run executable could not be found on the system path.
Searched for the executable 'xvfb-run' on the following path:
/usr/local/nvidia/bin
/usr/local/cuda/bin
/usr/local/sbin
/usr/local/bin
/usr/sbin
/usr/bin
My SetUp:Windows 10VS Code to run a Flask app
Google Colab to host the ML model
Python 3.9.7 used via Anaconda
I have noted several discussions where users recommend using Kaleido instead of Orca, and I am using Kaleido within the Flask app, however, I do not believe Colab supports Kaleido. Is this correct?
werruww
commented
3 days ago
I am receiving a Path Error using Orca in Google Colab to save an image in a Google Colab .ipynb Python Machine Learning model. A Flask App (written in VS Code IDE) uses the Colab trained model to output results to a URL. To be clear, the errors occur in my Colab Python script. This error is very similar to issue #252, however, I am unable to recreate the solution used there.
I Colab, I have created the following recommended import lines, as described in many posts:
!pip install plotly>=4.7.1 !wget https://github.com/plotly/orca/releases/download/v1.2.1/orca-1.2.1-x86_64.AppImage -O /usr/local/bin/orca !chmod +x /usr/local/bin/orcaandimport plotly.graph_objects as goI then use the following image write command:
fig.write_image('base_pic.svg')I expect to see 'base_pic.svg' appear as a file when the code runs, but this is not happening.My error seems to related to a path setting, however, I am unable to find detailed instruction to update and correct this path. Unfortunately, I am quite novice with troubleshooting paths.
The Error:
My SetUp:
Windows 10VS Codeto run a Flask appGoogle Colabto host the ML modelPython 3.9.7used via AnacondaI have noted several discussions where users recommend using Kaleido instead of Orca, and I am using Kaleido within the Flask app, however, I do not believe Colab supports Kaleido. Is this correct?
Any help would be welcomed. Mark