Q: export dataset as json

[SauravMaheshkar]

What is the best way to convert a weave dataset to a JSON or a Python sequential type?

At the moment, I use the following snippet as a template:

import json

def export_dataset_as_json(ref: str):
  vals = weave.ref(ref).get().rows.rows
  d = []
  for row in vals:
    d.append(dict(row))
  return json.dumps(d)

export_dataset_as_json(
    ref = "weave:///ml-colabs/ml-frameworks-bot/object/keras_evaluation_dataset:JgPCHXYZcQYjuaVzSRCvrBaFZO1yFM8saj1bUp5zrhw"
)

Feature Request

Something like a to_json() function or add_row() would be ideal. I primarily use this when I have to update a dataset. Happy to open a PR for this 😄

[gtarpenning]

Hey @SauravMaheshkar, thanks for writing in. Dataset manipulation in python (and in the UI!) is on our internal roadmap. In the meantime I am happy to review a PR if you make one, with the small caveat that the team might end up changing the design when we add a more comprehensive database experience. If you run into any blocking issues i'll be here to assist.