comproscanner 0.1.0__py3-none-any.whl

comproscanner/__init__.py

@@ -0,0 +1,182 @@
+"""
+ComProScanner - A package for extracting composition-property data from scientific articles.
+
+This package provides tools to collect metadata, process articles from various sources,
+extract composition-property relationships, evaluate extraction performance and visualize data distribution.
+
+Main functions:
+- collect_metadata: Collect and filter metadata from scientific articles
+- process_articles: Process articles from various sources (Elsevier, Wiley, etc.)
+- extract_composition_property_data: Extract composition-property relationships from articles
+- evaluate_semantic: Evaluate extraction quality using semantic similarity
+- evaluate_agentic: Evaluate extraction quality using agent-based methods
+- create_knowledge_graph: Create knowledge graph from extracted data
+"""
+
+# Import the main class
+from .comproscanner import ComProScanner
+
+# Import core configuration classes
+from .utils.configs.rag_config import RAGConfig
+from .utils.configs.llm_config import LLMConfig
+
+# Import visualization module
+from . import eval_visualizer
+from . import data_visualizer
+
+# Package version
+__version__ = "0.1.0"
+
+# Importing options for "from comproscanner import *"
+__all__ = [
+    "ComProScanner",
+    "collect_metadata",
+    "process_articles",
+    "extract_composition_property_data",
+    "evaluate_semantic",
+    "evaluate_agentic",
+    "create_knowledge_graph",
+    "RAGConfig",
+    "LLMConfig",
+    "eval_visualizer",
+    "data_visualizer",
+]
+
+
+def collect_metadata(
+    main_property_keyword,
+    base_queries=None,
+    extra_queries=None,
+    start_year=None,
+    end_year=None,
+):
+    """
+    Collect metadata from scientific articles.
+
+    Args:
+        main_property_keyword (str): The main property keyword to search for
+        base_queries (list, optional): List of base queries to search for
+        extra_queries (list, optional): List of extra queries to search for
+        start_year (int, optional): Start year for the search
+        end_year (int, optional): End year for the search
+    """
+    scanner = ComProScanner(main_property_keyword=main_property_keyword)
+    return scanner.collect_metadata(
+        base_queries=base_queries,
+        extra_queries=extra_queries,
+        start_year=start_year,
+        end_year=end_year,
+    )
+
+
+def process_articles(
+    main_property_keyword,
+    property_keywords=None,
+    source_list=["elsevier", "wiley", "iop", "springer", "pdfs"],
+    **kwargs,
+):
+    """
+    Process articles for the main property keyword.
+
+    Args:
+        main_property_keyword (str): The main property keyword to search for
+        property_keywords (dict): Dictionary of property keywords for filtering
+        source_list (list, optional): List of sources to process
+        **kwargs: Additional arguments to pass to the process_articles method
+    """
+    scanner = ComProScanner(main_property_keyword=main_property_keyword)
+    return scanner.process_articles(
+        property_keywords=property_keywords, source_list=source_list, **kwargs
+    )
+
+
+def extract_composition_property_data(
+    main_property_keyword, main_extraction_keyword=None, **kwargs
+):
+    """
+    Extract composition-property data from articles.
+
+    Args:
+        main_property_keyword (str): The main property keyword
+        main_extraction_keyword (str): The main keyword to extract data for
+        **kwargs: Additional arguments to pass to the extract_composition_property_data method
+    """
+    scanner = ComProScanner(main_property_keyword=main_property_keyword)
+    return scanner.extract_composition_property_data(
+        main_extraction_keyword=main_extraction_keyword, **kwargs
+    )
+
+
+def evaluate_semantic(
+    ground_truth_file=None,
+    test_data_file=None,
+    weights=None,
+    output_file="semantic_evaluation_result.json",
+    extraction_agent_model_name="gpt-4o-mini",
+    is_synthesis_evaluation=True,
+    use_semantic_model=True,
+    primary_model_name="thellert/physbert_cased",
+    fallback_model_name="all-mpnet-base-v2",
+    similarity_thresholds=None,
+):
+    """
+    Evaluate the extracted data using semantic evaluation.
+
+    Args:
+        ground_truth_file (str, optional): Path to the ground truth file. Defaults to None.
+        test_data_file (str, optional): Path to the test data file. Defaults to None.
+        weights (dict, optional): Weights for the evaluation metrics. Defaults to None.
+        output_file (str, optional): Path to the output file for saving the evaluation results. Defaults to "semantic_evaluation_result.json".
+        extraction_agent_model_name (str, optional): Name of the agent model used for extraction. Defaults to "GPT-4o-mini".
+        is_synthesis_evaluation (bool, optional): A flag to indicate if synthesis evaluation is required. Defaults to True.
+        use_semantic_model (bool, optional): A flag to indicate if semantic model should be used for evaluation. Defaults to True.
+        primary_model_name (str, optional): Name of the primary model for semantic evaluation. Defaults to "thellert/physbert_cased".
+        fallback_model_name (str, optional): Name of the fallback model for semantic evaluation. Defaults to "all-mpnet-base-v2".
+        similarity_thresholds (dict, optional): Similarity thresholds for evaluation. Defaults to 0.8 for each metric.
+    """
+    scanner = ComProScanner(main_property_keyword="placeholder")
+    return scanner.evaluate_semantic(
+        ground_truth_file=ground_truth_file,
+        test_data_file=test_data_file,
+        weights=weights,
+        output_file=output_file,
+        agent_model_name=extraction_agent_model_name,
+        is_synthesis_evaluation=is_synthesis_evaluation,
+        use_semantic_model=use_semantic_model,
+        primary_model_name=primary_model_name,
+        fallback_model_name=fallback_model_name,
+        similarity_thresholds=similarity_thresholds,
+    )
+
+
+def evaluate_agentic(
+    ground_truth_file=None,
+    test_data_file=None,
+    output_file="detailed_evaluation.json",
+    extraction_agent_model_name="gpt-4o-mini",
+    is_synthesis_evaluation=True,
+    weights=None,
+    llm=None,
+):
+    """
+    Evaluate the extracted data using agentic evaluation.
+
+    Args:
+        ground_truth_file (str, optional): Path to the ground truth file. Defaults to None.
+        test_data_file (str, optional): Path to the test data file. Defaults to None.
+        output_file (str, optional): Path to the output file for saving the evaluation results. Defaults to "detailed_evaluation.json".
+        extraction_agent_model_name (str, optional): Name of the agent model used for extraction. Defaults to "GPT-4o-mini".
+        is_synthesis_evaluation (bool, optional): A flag to indicate if synthesis evaluation is required. Defaults to True.
+        weights (dict, optional): Weights for the evaluation metrics. Defaults to None.
+        llm (LLM, optional): An instance of the LLM class. Defaults to None.
+    """
+    scanner = ComProScanner(main_property_keyword="placeholder")
+    return scanner.evaluate_agentic(
+        ground_truth_file=ground_truth_file,
+        test_data_file=test_data_file,
+        output_file=output_file,
+        agent_model_name=extraction_agent_model_name,
+        is_synthesis_evaluation=is_synthesis_evaluation,
+        weights=weights,
+        llm=llm,
+    )