Skip to content

Hypergraph is more powerful. Transform unstructured text into structured knowledge with LLMs. Graphs, hypergraphs, and spatio-temporal extractions — with one command.

feat(model): 为AutoModel添加可视化展示功能

yifanfeng977b6210562bee
diff --git a/examples/auto_model_demo.py b/examples/auto_model_demo.py
index 9cd4ece..dfb18d2 100644
--- a/examples/auto_model_demo.py
+++ b/examples/auto_model_demo.py
@@ -253,6 +253,8 @@ def main():
except Exception as e:
print(f"⚠️ 回答失败: {e}\n")
+ a_document_model.show(label_extractor=lambda x: x.title)
+
print("\n" + "=" * 80)
print("示例运行完成!")
print("=" * 80)
diff --git a/hyperextract/templates/zh/general/compliance_logic.py b/hyperextract/templates/zh/general/compliance_logic.py
index 4c32199..51cf98d 100644
--- a/hyperextract/templates/zh/general/compliance_logic.py
+++ b/hyperextract/templates/zh/general/compliance_logic.py
@@ -132,15 +132,15 @@ class ComplianceLogic(AutoHypergraph[ComplianceEntity, ComplianceRule]):
top_k_for_search: 语义检索返回的节点/边数量,默认为 3
top_k_for_chat: 问答使用的节点/边数量,默认为 3
"""
- def nodeLabelExtractor(node: ComplianceEntity) -> str:
+ def node_label_extractor(node: ComplianceEntity) -> str:
return f"{node.name} ({node.type})"
- def edgeLabelExtractor(edge: ComplianceRule) -> str:
+ def edge_label_extractor(edge: ComplianceRule) -> str:
return edge.ruleType
super().show(
- node_label_extractor=nodeLabelExtractor,
- edge_label_extractor=edgeLabelExtractor,
+ node_label_extractor=node_label_extractor,
+ edge_label_extractor=edge_label_extractor,
top_k_nodes_for_search=top_k_for_search,
top_k_edges_for_search=top_k_for_search,
top_k_nodes_for_chat=top_k_for_chat,
diff --git a/hyperextract/templates/zh/general/encyclopedia_item.py b/hyperextract/templates/zh/general/encyclopedia_item.py
index 0d8dbd1..0b4495a 100644
--- a/hyperextract/templates/zh/general/encyclopedia_item.py
+++ b/hyperextract/templates/zh/general/encyclopedia_item.py
@@ -85,3 +85,22 @@ class EncyclopediaItem(AutoModel[EncyclopediaInfo]):
verbose=verbose,
**kwargs,
)
+
+ def show(
+ self,
+ *,
+ top_k: int = 3,
+ ):
+ """
+ 展示百科条目。
+
+ Args:
+ top_k: 问答使用的条目数量,默认为 3
+ """
+ def label_extractor(item: EncyclopediaInfo) -> str:
+ return f"{item.title} ({item.entityType})"
+
+ super().show(
+ label_extractor=label_extractor,
+ top_k=top_k,
+ )
diff --git a/hyperextract/templates/zh/general/operational_procedure.py b/hyperextract/templates/zh/general/operational_procedure.py
index adff52a..2571bff 100644
--- a/hyperextract/templates/zh/general/operational_procedure.py
+++ b/hyperextract/templates/zh/general/operational_procedure.py
@@ -138,17 +138,17 @@ class OperationalProcedure(AutoGraph[ProcedureStep, ProcedureTransition]):
top_k_for_search: 语义检索返回的节点/边数量,默认为 3
top_k_for_chat: 问答使用的节点/边数量,默认为 3
"""
- def nodeLabelExtractor(node: ProcedureStep) -> str:
+ def node_label_extractor(node: ProcedureStep) -> str:
return f"{node.stepId}: {node.name}"
- def edgeLabelExtractor(edge: ProcedureTransition) -> str:
+ def edge_label_extractor(edge: ProcedureTransition) -> str:
if edge.condition:
return edge.condition
return "→"
super().show(
- node_label_extractor=nodeLabelExtractor,
- edge_label_extractor=edgeLabelExtractor,
+ node_label_extractor=node_label_extractor,
+ edge_label_extractor=edge_label_extractor,
top_k_nodes_for_search=top_k_for_search,
top_k_edges_for_search=top_k_for_search,
top_k_nodes_for_chat=top_k_for_chat,
diff --git a/hyperextract/templates/zh/general/penalty_mapping.py b/hyperextract/templates/zh/general/penalty_mapping.py
index c409f05..ded94f7 100644
--- a/hyperextract/templates/zh/general/penalty_mapping.py
+++ b/hyperextract/templates/zh/general/penalty_mapping.py
@@ -129,15 +129,15 @@ class PenaltyMapping(AutoGraph[PenaltyNode, PenaltyPath]):
top_k_for_search: 语义检索返回的节点/边数量,默认为 3
top_k_for_chat: 问答使用的节点/边数量,默认为 3
"""
- def nodeLabelExtractor(node: PenaltyNode) -> str:
+ def node_label_extractor(node: PenaltyNode) -> str:
return f"{node.name} ({node.type})"
- def edgeLabelExtractor(edge: PenaltyPath) -> str:
+ def edge_label_extractor(edge: PenaltyPath) -> str:
return edge.relation
super().show(
- node_label_extractor=nodeLabelExtractor,
- edge_label_extractor=edgeLabelExtractor,
+ node_label_extractor=node_label_extractor,
+ edge_label_extractor=edge_label_extractor,
top_k_nodes_for_search=top_k_for_search,
top_k_edges_for_search=top_k_for_search,
top_k_nodes_for_chat=top_k_for_chat,
diff --git a/hyperextract/templates/zh/general/personal_profile.py b/hyperextract/templates/zh/general/personal_profile.py
index a26f91f..f10a3f8 100644
--- a/hyperextract/templates/zh/general/personal_profile.py
+++ b/hyperextract/templates/zh/general/personal_profile.py
@@ -91,3 +91,24 @@ class PersonalProfile(AutoModel[PersonalInfo]):
verbose=verbose,
**kwargs,
)
+
+ def show(
+ self,
+ *,
+ top_k: int = 3,
+ ):
+ """
+ 展示个人档案。
+
+ Args:
+ top_k: 问答使用的条目数量,默认为 3
+ """
+ def label_extractor(item: PersonalInfo) -> str:
+ if item.birthDate and item.deathDate:
+ return f"{item.name} ({item.birthDate}-{item.deathDate})"
+ return item.name
+
+ super().show(
+ label_extractor=label_extractor,
+ top_k=top_k,
+ )
diff --git a/hyperextract/templates/zh/general/regulation_profile.py b/hyperextract/templates/zh/general/regulation_profile.py
index ae4ecd5..0a81aa8 100644
--- a/hyperextract/templates/zh/general/regulation_profile.py
+++ b/hyperextract/templates/zh/general/regulation_profile.py
@@ -89,3 +89,24 @@ class RegulationProfile(AutoModel[RegulationInfo]):
verbose=verbose,
**kwargs,
)
+
+ def show(
+ self,
+ *,
+ top_k: int = 3,
+ ):
+ """
+ 展示规章元数据快照。
+
+ Args:
+ top_k: 问答使用的条目数量,默认为 3
+ """
+ def label_extractor(item: RegulationInfo) -> str:
+ if item.version:
+ return f"{item.title} (v{item.version})"
+ return item.title
+
+ super().show(
+ label_extractor=label_extractor,
+ top_k=top_k,
+ )
diff --git a/hyperextract/types/model.py b/hyperextract/types/model.py
index f618a62..81109b9 100644
--- a/hyperextract/types/model.py
+++ b/hyperextract/types/model.py
@@ -2,12 +2,13 @@
from pathlib import Path
from datetime import datetime
-from typing import List, Any, Type
+from typing import List, Any, Type, Callable
from ontomem.merger import MergeStrategy, create_merger, BaseMerger
from langchain_core.language_models.chat_models import BaseChatModel
from langchain_core.embeddings import Embeddings
from langchain_core.documents import Document
from langchain_community.vectorstores import FAISS
+from ontosight import view_nodes
from .base import BaseAutoType, T
from hyperextract.utils.logging import logger
@@ -296,6 +297,52 @@ class AutoModel(BaseAutoType[T]):
str(folder), self.embedder, allow_dangerous_deserialization=True
)
+ def show(
+ self,
+ label_extractor: Callable[[T], str],
+ *,
+ top_k: int = 3,
+ ) -> None:
+ """Visualize the model using OntoSight.
+
+ Args:
+ label_extractor: A function that takes a model instance and returns a string label for visualization.
+ top_k: Number of items to retrieve for chat callback (default: 3).
+ """
+
+ if self._index is not None:
+ logger.info(
+ "Visualizing model with search and chat capabilities (indices detected)."
+ )
+
+ def chat_callback(question: str) -> None:
+ response = self.chat(question, top_k=top_k)
+ content = response.content
+ retrieved_items = [self.data]
+ return content, retrieved_items
+ else:
+ logger.info(
+ "Visualizing list without search and chat capabilities (no indices detected)."
+ )
+ chat_callback = None
+
+ from hashlib import md5
+
+ def item_id_extractor(item: T) -> str:
+ return md5(str(item.model_dump()).encode()).hexdigest()[:8]
+
+ view_nodes(
+ node_list=[self.data],
+ node_schema=self._data_schema,
+ node_id_extractor=item_id_extractor,
+ node_label_extractor=label_extractor,
+ on_chat=chat_callback,
+ context={
+ "title": f"{self._data_schema.__name__} Model",
+ "description": f"Visualizing {len(self._data_schema.model_fields)} fields in AutoModel",
+ },
+ )
+
# ==================== Operators ====================
def __add__(self, other):