Hey everyone! I'm working on a modular pipeline and I'm trying to freeze one of my parameter inputs which is a dictionary:

features:
  numeric:
    x: "x"
  categorical:
    y:
      col: "y"
      dropna: True
    z:
      col: "z"
      dropna: True
    i:
      col: "i"
      dropna: False
    j:
      col: "j"
      dropna: False

but when I try to freeze the parameter:

ingestion_pipeline = pipeline(
    pipe=ingestion_pipe,
    inputs={
        "a",
        "b",
        "c",
        "d"
    },
    parameters="features",
    namespace="ingestion"
)

I get

Failed to map datasets and/or parameters: params:features

When I namespace

features

it works, Am I doing something wrong? I'm using

kedro 0.18.3

with spark