science_live.pipeline

`science_live.pipeline`#

Science Live Pipeline: 7-Step Natural Language Processing Pipeline#

This package implements a 7-step pipeline for processing natural language questions and converting them to structured queries against nanopub networks:

QuestionProcessor - Parse and classify questions
EntityExtractorLinker - Extract and link entities
RosettaStatementGenerator - Generate Rosetta statements
SPARQLGenerator - Convert to SPARQL queries
QueryExecutor - Execute queries
ResultProcessor - Structure results
NaturalLanguageGenerator - Generate natural language responses

Usage: # Individual components from science_live.pipeline import QuestionProcessor processor = QuestionProcessor()

# Complete pipeline
from science_live.pipeline import ScienceLivePipeline
pipeline = ScienceLivePipeline(endpoint_manager)
result = await pipeline.process("Your question here")

# Custom pipeline
from science_live.pipeline import create_custom_pipeline
custom_pipeline = create_custom_pipeline([
    QuestionProcessor(),
    EntityExtractorLinker(endpoint_manager),
    # ... other steps
])

Submodules#

Package Contents#

Functions#

`get_pipeline_status`	Get status of pipeline components
`event_loop`	Create an instance of the default event loop for the test session.
`processing_context`	Create a basic processing context for tests.
`mock_endpoint_manager`	Create a mock endpoint manager for testing.
`sample_extracted_entities`	Create sample extracted entities for testing.
`sample_sparql_results`	Create sample SPARQL results for testing.
`pytest_configure`	Configure pytest with custom markers.
`pytest_runtest_setup`	Skip tests that require unavailable components.

Data#

`_COMPONENTS_AVAILABLE`
`__all__`
`__version__`
`__author__`
`__description__`

API#

science_live.pipeline._COMPONENTS_AVAILABLE = None#

science_live.pipeline.__all__ = []#

science_live.pipeline.get_pipeline_status()#: Get status of pipeline components

science_live.pipeline.__version__ = '1.0.0'#

science_live.pipeline.__author__ = 'Science Live Team'#

science_live.pipeline.__description__ = 'Semantic knowledge exploration for scientific research'#

science_live.pipeline.event_loop()#: Create an instance of the default event loop for the test session.

science_live.pipeline.processing_context()#: Create a basic processing context for tests.

science_live.pipeline.mock_endpoint_manager()#: Create a mock endpoint manager for testing.

science_live.pipeline.sample_extracted_entities()#: Create sample extracted entities for testing.

science_live.pipeline.sample_sparql_results()#: Create sample SPARQL results for testing.

science_live.pipeline.pytest_configure(config)#: Configure pytest with custom markers.

science_live.pipeline.pytest_runtest_setup(item)#: Skip tests that require unavailable components.