from sotastream.augmentors import DataSource, UTF8File
from . import Pipeline, pipeline
[docs]
@pipeline('default')
class DefaultPipeline(Pipeline):
def __init__(self, parallel_data, **kwargs):
super().__init__(**kwargs)
self.stream = self.create_data_stream(parallel_data)
[docs]
@classmethod
def get_data_sources_for_argparse(cls):
return [('parallel_data', 'Path to parallel data (folder with .gz files, or compressed TSV)')]
[docs]
@classmethod
def get_data_sources_default_weights(cls):
return [1.0]