You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
 
 
 
 

85 lines
3.0 KiB

from . import builders, catalog, cleaners, data, labels, parsers, readers
from projects.models import (
DOCUMENT_CLASSIFICATION,
IMAGE_CLASSIFICATION,
INTENT_DETECTION_AND_SLOT_FILLING,
SEQ2SEQ,
SEQUENCE_LABELING,
SPEECH2TEXT,
)
def get_data_class(project_type: str):
text_projects = [DOCUMENT_CLASSIFICATION, SEQUENCE_LABELING, SEQ2SEQ, INTENT_DETECTION_AND_SLOT_FILLING]
if project_type in text_projects:
return data.TextData
else:
return data.FileData
def create_parser(file_format: str, **kwargs):
mapping = {
catalog.TextFile.name: parsers.TextFileParser,
catalog.TextLine.name: parsers.LineParser,
catalog.CSV.name: parsers.CSVParser,
catalog.JSONL.name: parsers.JSONLParser,
catalog.JSON.name: parsers.JSONParser,
catalog.FastText.name: parsers.FastTextParser,
catalog.Excel.name: parsers.ExcelParser,
catalog.CoNLL.name: parsers.CoNLLParser,
catalog.ImageFile.name: parsers.PlainParser,
catalog.AudioFile.name: parsers.PlainParser,
}
if file_format not in mapping:
raise ValueError(f"Invalid format: {file_format}")
return mapping[file_format](**kwargs)
def get_label_class(project_type: str):
mapping = {
DOCUMENT_CLASSIFICATION: labels.CategoryLabel,
SEQUENCE_LABELING: labels.SpanLabel,
SEQ2SEQ: labels.TextLabel,
IMAGE_CLASSIFICATION: labels.CategoryLabel,
SPEECH2TEXT: labels.TextLabel,
}
if project_type not in mapping:
ValueError(f"Invalid project type: {project_type}")
return mapping[project_type]
def create_cleaner(project):
mapping = {
DOCUMENT_CLASSIFICATION: cleaners.CategoryCleaner,
SEQUENCE_LABELING: cleaners.SpanCleaner,
IMAGE_CLASSIFICATION: cleaners.CategoryCleaner,
}
if project.project_type not in mapping:
return cleaners.Cleaner(project)
cleaner_class = mapping.get(project.project_type, cleaners.Cleaner)
return cleaner_class(project)
def create_builder(project, **kwargs):
if not project.is_text_project:
return builders.PlainBuilder(data_class=get_data_class(project.project_type))
data_column = builders.DataColumn(
name=kwargs.get("column_data") or readers.DEFAULT_TEXT_COLUMN, value_class=get_data_class(project.project_type)
)
# If project is intent detection and slot filling,
# column names are fixed: entities, cats
if project.project_type == INTENT_DETECTION_AND_SLOT_FILLING:
label_columns = [
builders.LabelColumn(name="cats", value_class=labels.CategoryLabel),
builders.LabelColumn(name="entities", value_class=labels.SpanLabel),
]
else:
label_columns = [
builders.LabelColumn(
name=kwargs.get("column_label") or readers.DEFAULT_LABEL_COLUMN,
value_class=get_label_class(project.project_type),
)
]
builder = builders.ColumnBuilder(data_column=data_column, label_columns=label_columns)
return builder