def publish_clusters(session: Session, project: MasteringProject) -> Operation:
"""Publish current record clusters and wait for the operation to complete
Args:
project: Tamr Mastering project
"""
op = _publish_clusters_async(session, project)
return operation.wait(session, op)
def estimate_pairs(session: Session, project: MasteringProject) -> Operation:
"""Update the estimated pair counts and wait for the operation to complete
Args:
project: Tamr Mastering project
"""
op = _estimate_pairs_async(session, project)
return operation.wait(session, op)
def update_unified_dataset(session: Session, project: MasteringProject) -> Operation:
"""Apply changes to the unified dataset and wait for the operation to complete
Args:
project: Tamr Mastering project
"""
op = _update_unified_dataset_async(session, project)
return operation.wait(session, op)
def publish(session: Session, project: GoldenRecordsProject) -> Operation:
"""Publish the golden records and wait for the operation to complete
Args:
project: Tamr Golden Records project
"""
op = _publish_async(session, project)
return operation.wait(session, op)
def update(session: Session, project: GoldenRecordsProject) -> Operation:
"""Update the draft golden records and wait for the operation to complete
Args:
project: Tamr Golden Records project
"""
op = _update_async(session, project)
return operation.wait(session, op)
def apply_changes(session: Session, unified_dataset: UnifiedDataset) -> Operation:
"""Applies changes to the unified dataset and waits for the operation to complete
Args:
unified_dataset: The Unified Dataset which will be committed
"""
op = _apply_changes_async(session, unified_dataset)
return operation.wait(session, op)
def update_results(session: Session, project: CategorizationProject) -> Operation:
"""Generate classifications based on the latest categorization model and wait for the
operation to complete
Args:
project: Tamr Categorization project
"""
op = _update_results_async(session, project)
return operation.wait(session, op)
def apply_feedback(session: Session, project: CategorizationProject) -> Operation:
"""Train the categorization model according to verified labels and wait for the
operation to complete
Args:
project: Tamr Categorization project
"""
op = _apply_feedback_async(session, project)
return operation.wait(session, op)
def update_cluster_results(session: Session, project: MasteringProject) -> Operation:
"""Generate clusters based on the latest pair-matching model and wait for the
operation to complete
Args:
project: Tamr Mastering project
"""
op = _update_cluster_results_async(session, project)
return operation.wait(session, op)
def update_high_impact_pairs(session: Session, project: MasteringProject) -> Operation:
"""Produce new high-impact pairs according to the latest pair-matching model and
wait for the operation to complete
Args:
project: Tamr Mastering project
"""
op = _update_high_impact_pairs_async(session, project)
return operation.wait(session, op)
def update_pair_results(session: Session, project: MasteringProject) -> Operation:
"""Update record pair predictions according to the latest pair-matching model and
wait for the operation to complete
Args:
project: Tamr Mastering project
"""
op = _update_pair_results_async(session, project)
return operation.wait(session, op)
def apply_feedback(session: Session, project: MasteringProject) -> Operation:
"""Train the pair-matching model according to verified labels and wait for the
operation to complete
Args:
project: Tamr Mastering project
"""
op = _apply_feedback_async(session, project)
return operation.wait(session, op)
def generate_pairs(session: Session, project: MasteringProject) -> Operation:
"""Generate pairs according to the binning model and wait for the operation
to complete
Args:
project: Tamr Mastering project
"""
op = _generate_pairs_async(session, project)
return operation.wait(session, op)
def materialize(session: Session, dataset: Dataset) -> Operation:
"""Materialize a dataset and wait for the operation to complete
Materializing consists of updating the dataset (including records) in persistent storage (HBase) based on upstream changes to data.
Args:
dataset: A Tamr dataset which will be materialized
"""
op = _materialize_async(session, dataset)
return operation.wait(session, op)
def update_unified_dataset(
session: Session, project: CategorizationProject
) -> Operation:
"""Apply changes to the unified dataset and wait for the operation to complete
Args:
project: Tamr Categorization project
"""
unified_dataset = unified.from_project(session, project)
op = unified._apply_changes_async(session, unified_dataset)
return operation.wait(session, op)