feat: add General Functions to TensorFlow frontend #27083
Conversation
![github-actions[bot]](https://avatars.githubusercontent.com/in/15368?s=60&v=4){kind=small}
There was a problem hiding this comment.
PR Compliance Checks
Thank you for your Pull Request! We have run several checks on this pull request in order to make sure it's suitable for merging into this project. The results are listed in the following section.
Protected Branch
In order to be considered for merging, the pull request changes must not be implemented on the "main" branch. This is described in our Contributing Guide. We are closing this pull request and we would suggest that you implement your changes as described in our Contributing Guide and open a new pull request.
ivy-leaves
added
the
TensorFlow Frontend
There was a problem hiding this comment.
Congrats on making your first Pull Request and thanks for supporting Ivy! 🎉
Join the conversation in our Discord.
For every PR opened, we run unit tests and comment the results in the PR to ensure the functionality remains intact.
Please make sure they are passing. 💪
If you want to check them from the action runners, you can open the display_test_results job. 👀
It contains the following two sections:
- Combined Test Results: This shows the results of all the ivy tests that ran on the PR. ✔️
- New Failures Introduced: This lists the tests that fail on this PR.
Keep in mind that we will assign an engineer for this task and they will look at it based on the workload that they have, kindly be patient 😄.
|
Thank you for this PR, here is the CI results: This pull request does not result in any additional test failures. Congratulations! |
There was a problem hiding this comment.
@pixel3user looking at the tensorflow source code for this (https://github.com/tensorflow/tensorflow/blob/v2.14.0/tensorflow/python/ops/control_flow_ops.py#L1957-L2033), the function does more than just no_op. But not exactly sure which parts of the functionality we will need to replicate within our frontend. For example, do we need to replicate the behaviour of control dependencies which the docs note are not really relevant as of tensorflow 2 (https://www.tensorflow.org/api_docs/python/tf/group). Maybe @AnnaTz has some thoughts on this?
pixel3user
changed the title
|
As TensorFlow 2 have automatic control dependencies than maybe we don't need to replicate the functionality in the frontend functions. I am looking forward to have a response from you. |
|
Hi @Sam-Armstrong @pixel3user, currently we are not looking to alter the graph execution from the frontend level. And, if it becomes desirable in the future, we would need to implement dedicated ivy functions for this purpose first. |
|
Sure makes sense, sorry that this is the outcome @pixel3user, feel free to pick up another of the todo list issues if you'd like 😊 |
feat: add group function to Ivy TensorFlow frontend
Add General Functions to TensorFlow frontend #6931
Closes #27080
Checklist
Did you add a function?
Did you add the tests?
Did you run your tests and are your tests passing?
Did pre-commit not fail on any check?
Did you follow the steps we provided?
Socials