-
Notifications
You must be signed in to change notification settings - Fork 720
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Missing tfx.orchestration in Colab tutorials #5610
Comments
Hi, @michaelwsherman Apologize for the inconvenience and Yes, currently we're experiencing that issue with orchestration and our team is making some code level changes so once it's done issue will be resolved and I'll update the same here once I got the confirmation from our concerned team about issue resolution Thank you! |
Hi, @michaelwsherman I was able to replicate the issue without error and for that you'll have to uninstall Could you please try and confirm, Is it working both the tutorials as expected from your end ? Could you please confirm if this issue is resolved for you ? Please feel free to close the issue if it is resolved ? Thank you! |
Can confirm both gists work, restarting where instructed to. Note this applies to more of the TFX tutorials than these two, I just stopped checking after these two. I'd suggest fixing the others before resolving this issue. |
Hi, @michaelwsherman I checked all the TFX tutorials under the Tutorials section and they're working as expected after uninstalling @lego0901, I found one tutorial where new changes are not reflected so could you please look into this tutorial where uninstall shapely section is missing ? Thank you! |
Thanks for catching the missing notebook! I think PR you made will solve the issue correctly! And I guess we should wait for the TensorFlow folks to response. I appreciate your consideration! |
Hi, @lego0901 You're welcome and yes, I have submitted PR for that so we'll wait for response from TensorFlow folks. Thank you! |
This issue has been marked stale because it has no recent activity since 14 days. It will be closed if no further activity occurs. Thank you. |
System information
pip freeze
output): This is massive, I've put it at the bottom.Describe the current behavior
In TFX tutorial colabs, when I run the code:
tfx.orchestration.LocalDagRunner().run(whatever)
I get:
AttributeError: module 'tfx.v1' has no attribute 'orchestration'
I've seen this in two separate TFX tutorial notebooks: penguin_simple.ipynb and penguin_tfdv.ipynb. I am following the instructions in the notebook to restart the runtime after pipping TFX.
I've experienced the issue in penguin_simple.ipynb using colab on two separate Google accounts. In my own local dev environment, I can run the code from penguin_simple.ipynb and it works. In both colab and my own environment,
print('TFX version: {}'.format(tfx.__version__))
outputsTFX version: 1.12.0
.Investigating a bit, in colab
dir(tfx)
shows noorchestration
, but in my own environmentorchestration
is present.My environment is Python 3.7.10 but colab is Python version 3.8.16, maybe this is related?
Describe the expected behavior
The notebooks should work in colab.
Standalone code to reproduce the issue
Not quite standalone code, but if you go to the Simple TFX Pipeline Tutorial page and open the Run in Google Colab link, then run through the notebook (including restarting after
pip install -U tfx
), you'll get this error on the second to last code block.pip freeze:
The text was updated successfully, but these errors were encountered: