Reformat gradients

[saeedashrraf]

As explained in the function reformatting open task, each function in the gradients submodule should be updated to follow the implementation instructions given in the Deep Dive section of the docs. The updates should be applied for both the Ivy API and Backend API. The Deep Dive is the best general resource for learning in detail how functions should be implemented. Many of the functions are not implemented as they should be. You will need to make changes to the current implementations, but you do not need to go through all section of the Deep Dive. Specifically, you should make the following changes, where appropriate:

1. remove all lambda and direct bindings for backend functions, with each function instead defined using def.
2. update the function arguments and type hints.
3. add the correct docstrings
4. add thorough docstring examples

_

Please keep in mind that the proper way to link an issue to this list is to comment "- [ ] #issue_number" while the issue's title only includes the name of the function you've chosen.

_

• with_grads #1775
• set_with_grads #27103
• unset_with_grads #27124
• variable #2323
• is_variable #1799
• variable_data #9283
• stop_gradient #2241
• Reformat execute_with_gradients #10655
• adam_step #1640
• optimizer_update #1833
• gradient_descent_update #2143
• lars_update #25688
• adam_update #25949
• lamb_update

The main file paths where these functions are likely to be added are:

• ivy/functional/ivy/gradients.py
• ivy/array/gradients.py
• ivy/container/gradients.py
• ivy/functional/backends/jax/gradients.py
• ivy/functional/backends/tensorflow/gradients.py
• ivy/functional/backends/torch/gradients.py

[a0m0rajab]

If we look at the check list it seems like all the functions here has been merged, but it seems that:

• set_with_grads
• lars_update
• adam_update
• lamb_update
  has no merged PRs, are they fixed or are they still open for contribution?
  @saeedashrraf

[TangoIndiaMango]

@a0m0rajab
according to the functions you posted, did you get a reply for a go ahead?

[a0m0rajab]

@TangoIndiaMango not yet, if I got something it would be written in this issue.

[rehanganapathy]

Hey I'd like to work on this issue if it hasn't been resolved yet

[tbeetech]

• set_with_grads

[tbeetech]

• lamb_update

[Janhavipadhi]

unset_with_grads #26865