Using namespace with viz without changing Kedro pipeline structure

[noklam]

Description

I want to be able to show my pipeline as a "collapsable" one on viz, currently only namespace pipeline is possible but it requires me to change how I name my catalog, parameters.

Context

Why is this change important to you? How would you use it? How can it benefit other users?
Namespace is a big change to Kedro pipeline structure and I don't necessary want to buy in to the whole modular pipeline structure. It should be possible to do this purely as a viz feature without changing my Kedro pipeline.

Alternative

Use the keys in pipeline_registry.py as the "subpipeline" for collapase/expand.

Attempt

This is my best attempt to keep dataset/parameters as is, but modular pipeline prevent me to rename intermediate dataset with validation error. How does kedro viz group things into a "namespace"? Is it the node.name attribute, if so how dataset and params are associated with it?

def create_pipeline(**kwargs) -> Pipeline:
    p= pipeline(
        [
            node(
                func=split_data,
                inputs=["model_input_table", "params:model_options"],
                outputs=["X_train", "X_test", "y_train", "y_test"],
                name="split_data_node",
            ),
            node(
                func=train_model,
                inputs=["X_train", "y_train"],
                outputs="regressor",
                name="train_model_node",
            ),
            node(
                func=evaluate_model,
                inputs=["regressor", "X_test", "y_test"],
                outputs=None,
                name="evaluate_model_node",
            ),

        ],

    )

    # attempt to escape from namespace so I don't have to rename all dataset
    i = {}
    o = {}
    pa = {}
    for inp in p.inputs():
        print("!!!!", inp)
        if isinstance(inp, dict):
            pass
        elif not inp.startswith("param"):
            i[inp] = inp
        else:
            pa[inp] = inp
    for out in p.outputs():
        o[out] = out
    p = pipeline(p, namespace="nok",inputs=i, outputs=o, parameters=pa)
    return p