Kedro is an open-sourced Python framework for creating maintainable and modular data science code.

Kedro

hello!
is there a way in kedro that I can pass a custom resolver which can be treated like `runtime_params` ?
for example,
```model_options:
  random_state: "${my_resolver:random, ${globals:my_global_value}}"```
will this be legit in kedro?

Hey <@U07EJ2XNWD8>, could you provide some extra context on what are you trying to implement?

Of course!
currently, if we have a `parameter.yaml` where we specify
```model_options:
  random_state: "${runtime_params:random, ${globals:my_global_value}}"```
then when we run
```kedro run --params random=3```
we can overwrite the value of `random_state` with what we provide at runtime, which is 3 in this case.

I want to make the following tweaks possible:
in `parameter.yaml` we write
```model_options:
  random_state: "${apple:random, ${globals:my_global_value}}"```
and if I provide
```kedro run --params random=3```
then, `random_state` will be overwritten by value 3 during runtime.

I also want kedro to recognise/understand that parameter prefixed by  `apple:` is also a overwritable runtime parameter