Is there a way flatten arrays referenced from external files in serverless.yml

serverless-python-requirements enables packaging large python libraries in layers and specifying them in noDeploy so as to avoid redeploying these large libraries along with the functions as shown here.

I have a file containing the layer arn and noDeploy for such libraries in a file and I would like to make every other lambda function refer to that file.

Example of such a file - python_req.yml -

pandas:
    arn: <---->
    noDeploy:
      - numpy
      - pandas
      - python-dateutil
      - pytz
      - six
requests:
    arn: <---->
    noDeploy:
      - certifi
      - chardet
      - idna
      - requests
      - urllib3

Assuming python_req.yml exists a level above a folder named my-sample with a function hello to which I have added the following layers

layers:
  - ${file(../python-req.yml):pandas.arn}
  - ${file(../python-req.yml):requests.arn}

I would like to also combine the noDeploy list of pandas and requests

custom:
  pythonRequirements:
    dockerizePip: non-linux
    noDeploy:
      - ${file(../python-req.yml):pandas.noDeploy}
      - ${file(../python-req.yml):requests.noDeploy}

doing this makes custom.pythonRequirements.noDeploy an array of arrays
[[‘numpy’, ‘pandas’, ‘python-dateutil’, ‘pytz’, ‘six’], [‘certifi’, ‘chardet’, ‘idna’, ‘requests’, ‘urllib3’]]

and serverless-python-requirements expects noDeploy to be an array and packages all these dependencies along with the function instead of ignoring them

Is there a way flatten arrays referenced from external file in serverless.yml