Problems with packages

#15
by Lion815 - opened

I used the codes below:\

from datasets import load_dataset

dataset = load_dataset('wikipedia', '20220301.aa')

however, there is an error:

File c:\Users\48139\.conda\envs\study\lib\site-packages\apache_beam\coders\coders.py:57, in 
     54 from typing import overload
     56 import google.protobuf.wrappers_pb2
---> 57 import proto
     59 from apache_beam.coders import coder_impl
     60 from apache_beam.coders.avro_record import AvroRecord
ModuleNotFoundError: No module named 'proto'

But I have already done

pip install apache_beam mwparserfromhell

and pip install proto did not work, what shall I do?

Datasets Maintainers org

Thanks for reporting, @Lion815 .

Note that we are going to deprecate this dataset: it only contains the pre-processed data for 6 of the languages.

I would recommend that you use the current official "wikimedia/wikipedia" dataset, with the pre-processed data for all the languages and for the latest dump 2023-11-01.

albertvillanova changed discussion status to closed

Sign up or log in to comment