archaeo_super_prompt.modeling

source package archaeo_super_prompt.modeling

Subpackage containing the models and the pipelines for the experiments.

Compose them to run inferences, training or evaluations as in the standard sklearn workflow.

Modules

• archaeo_super_prompt.modeling.categories

• archaeo_super_prompt.modeling.entity_extractor — Root of the module for infering in the NER model.

• archaeo_super_prompt.modeling.pdf_to_text — PDF Ingestion layer with vision llm and chunking model.

• archaeo_super_prompt.modeling.struct_extract

• archaeo_super_prompt.modeling.types

• archaeo_super_prompt.modeling.DAG_builder — Interface over the DAG builder of skdag.

• archaeo_super_prompt.modeling.chunk_selector — Utils to select boundaries page of interest in documents.

• archaeo_super_prompt.modeling.legacy_predict — Legacy model for comparison.

• archaeo_super_prompt.modeling.other_dag — DAGs to train the FieldExtractor models.

• archaeo_super_prompt.modeling.predict — Code containing the global model and a way to infer into it.

• archaeo_super_prompt.modeling.similarity

• archaeo_super_prompt.modeling.train — DAGs to train the FieldExtractor models.