This browser is no longer supported.
Upgrade to Microsoft Edge to take advantage of the latest features, security updates, and technical support.
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(128, 8%, 22%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EA%3C/text%3E%3C/svg%3E)
I have been using the Azure ML python SDK and it has been working fine until I started getting this error:
"ImportError: DLL load failed while importing win32file: The specified procedure could not be found." when I tried to access my workspace
I am using a conda virtual environment with Python 3.9 and I was running all my codes in a jupyter notebook on a windows computer.
It's been many hours now and I cannot find a solution. Can you help?
@Anonymous Is there any other module or library name listed in the error message where the import of this module failed when the procedure is called? Have you tried to uninstall and install this module or the SDK or restart the kernel and check if it helps?
I can't see any The thing is it was working fine (connection, creating a resource, workspace, etc.) and I didn't install any additional library afterward. I came back to resume my experiment and got welcomed by those errors.
I have attached all the details about the error. 187276-error.pdf
@Anonymous It looks like a localized issue while trying to call the interactive authentication method from the SDK where some of the core modules of windows are failing to import. Could you please try to restart your notebook service and load the notebook again and authenticate with your Azure workspace? I think with the current state of your setup even a call to print the SDK version would fail.
import azureml.core
print(azureml.core.VERSION)
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(144, 55.00000000000001%, 23%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EPM%3C/text%3E%3C/svg%3E)
I have very similar issue: local PC with Anaconda environment where I additionally installed Azure ML SDK.
Running just:
import azureml.core
print(azureml.core.VERSION)
works fine, but when I add to such file:
workspace = azureml.core.Workspace.from_config()
It will throw similar error:
https://pastebin.com/QzaNB7KA
Interestingly enough when I installed this SDK on my "regular" Python 3.6.8 environment (not Anaconda one) it seems to work fine, although there I don't have any ML-related libraries installed like PyTorch, Pandas, etc. so it could be nice to make it work with Anaconda as well...
This is definitely still an issue and Anaconda Individual Edition is definitely an environment that should be supported for data science workflows on Azure.
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(224.00000000000003, 53%, 31%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EMH%3C/text%3E%3C/svg%3E)
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(96, 93%, 19%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3ENV%3C/text%3E%3C/svg%3E)
I have the same issue (when using azureml.core.authentication.ServicePrincipalAuthentication) but the fix/es referred to in this answer did not work for me while using Python 3.9. After I switched to an environment that uses Python 3.7, it worked. Specifically, I installed pywin32 and ran pywin32-postinstall.