Dataset_id required on dataset creation

[joaquinvanschoren]

When I want to upload a new dataset, I can create an OpenMLDataset and then publish it.
However, the creation of an OpenMLDataset requires a dataset_id.

This does not make sense when uploading a new dataset. I could just add dataset_id='1', but that is still confusing since it will be entirely ignored.

[joaquinvanschoren]

My code. Remove dataset_id from the list of parameters to see the error.

oml.config.server = 'https://test.openml.org/api/v1'
my_data = oml.datasets.OpenMLDataset(data_file='data/ram_price.arff', name='RAM_Price', description='This is a test.', version='1', tag='test', format='ARFF', licence='Public', visibility='public', dataset_id='1')
response = my_data.publish()

[mfeurer]

@ArlindKadra could you please check whether this is solved by #440?

[ArlindKadra]

@mfeurer

openml-python/openml/datasets/dataset.py

Line 35 in 4603345

dataset_id : int, optional

[mfeurer]

Thanks @ArlindKadra