primeqa.qg.processors.table_qg.sql_processor.SqlProcessor#

class primeqa.qg.processors.table_qg.sql_processor.SqlProcessor(tokenizer: transformers.tokenization_utils.PreTrainedTokenizer, input_max_len: int, target_max_len: int)#

Bases: object

Class for sql dataset processing, like wikisql, contains methods to preprocess data execute sql etc.

Methods

convert_to_features

load_tables

preprocess_data

_summary_

Attributes

tokenizer

input_max_len

target_max_len

__call__(dataset) datasets.arrow_dataset.Dataset#

Call self as a function.

static preprocess_data(example_batch: Dict)#

_summary_

Parameters

example_batch (Dict) – _description_