You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When we working with cloud-native applications we often have a CLI interface. Would be great if we can run tensorflow-cloud in the terminal using something like tfc run mnist_example.py
Thank you for raising this Eddy, as a work around you could have your tfc.run() script in python file say my_example.py. Then from the command prompt you can run it by
No totally because the idea is that the user doesn't need to create a python file if he no wants to, instead of executing a command in the command line is enough, for example:
tfc run mnist_example.py
This is another interface to use tf-cloud. We can use tf-cloud with .py files, .ipynb files, and with this we will use tf-cloud through CLI.
Activity
SinaChavoshi commentedon Jan 20, 2021
Thank you for raising this Eddy, as a work around you could have your tfc.run() script in python file say my_example.py. Then from the command prompt you can run it by
see call_run_on_notebook_with_keras_fit.py for an example.
Kindly let me know if this address your use case.
eddydecena commentedon Feb 3, 2021
No totally because the idea is that the user doesn't need to create a python file if he no wants to, instead of executing a command in the command line is enough, for example:
This is another interface to use tf-cloud. We can use tf-cloud with .py files, .ipynb files, and with this we will use tf-cloud through CLI.
See pull request #283.
SinaChavoshi commentedon Feb 21, 2021
Thank you for the work Eddy, I left comments in the PR #283