Table Question Answering

Type: huggingface.question_answering.TableQuestionAnswering

Namespace: huggingface.question_answering

Description

Answers questions based on tabular data. table, question answering, natural language processing

Use cases:
- Querying databases using natural language
- Analyzing spreadsheet data with questions
- Extracting insights from tabular reports
- Automated data exploration

Properties

Property Type Description Default
model hf.table_question_answering The model ID to use for table question answering {'type': 'hf.table_question_answering', 'repo_id': '', 'path': None, 'variant': None, 'allow_patterns': None, 'ignore_patterns': None}
dataframe dataframe The input table to query {'type': 'dataframe', 'uri': '', 'asset_id': None, 'data': None, 'columns': None}
question str The question to be answered based on the table ``

Outputs

Output Type Description
answer str  
coordinates List[Tuple[int, int]]  
cells List[str]  
aggregator str  

Metadata

Browse other nodes in the huggingface.question_answering namespace.