Spaces:
Running
Running
File size: 9,199 Bytes
c2ea5ed f0fc928 c2ea5ed 7e807a3 c2ea5ed f0fc928 c2ea5ed f0fc928 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 |
"""
Method Registry for Knowledge Extraction Methods
This module provides a centralized registry for all available knowledge extraction methods
and their associated schemas. Each method is bound to a specific schema type.
"""
from enum import Enum
from typing import Any, Dict, List, Optional
class MethodType(Enum):
"""Types of extraction methods"""
PRODUCTION = "production"
BASELINE = "baseline"
class SchemaType(Enum):
"""Types of schemas used by methods"""
REFERENCE_BASED = "reference_based"
DIRECT_BASED = "direct_based"
# Method Registry - maps method names to their implementations and schemas
AVAILABLE_METHODS = {
# Production method: OpenAI Structured Outputs (simple, direct_call)
"openai_structured": {
"name": "OpenAI Structured Outputs",
"description": "Simple OpenAI structured outputs extractor using Pydantic models",
"method_type": MethodType.PRODUCTION,
"schema_type": SchemaType.REFERENCE_BASED,
"module_path": "agentgraph.methods.production.openai_structured_extractor",
"class_name": "OpenAIStructuredFactory",
"supported_features": ["structured_outputs", "direct_extraction"],
"processing_type": "direct_call"
},
# Production method using reference-based schema
"production": {
"name": "Multi-Agent Knowledge Extractor",
"description": "Production CrewAI-based multi-agent system with content reference resolution",
"method_type": MethodType.PRODUCTION,
"schema_type": SchemaType.REFERENCE_BASED,
"module_path": "agentgraph.methods.production.multi_agent_knowledge_extractor",
"class_name": "agent_monitoring_crew_factory",
"supported_features": ["content_references", "failure_detection", "line_numbers"],
"processing_type": "async_crew"
},
"openai_structured": {
"name": "OpenAI Structured Outputs",
"description": "Simple OpenAI structured outputs extractor using Pydantic models",
"method_type": MethodType.PRODUCTION,
"schema_type": SchemaType.REFERENCE_BASED,
"module_path": "agentgraph.methods.production.openai_structured_extractor",
"class_name": "OpenAIStructuredFactory",
"supported_features": ["structured_outputs", "direct_extraction"],
"processing_type": "direct_call"
},
# Baseline methods using direct-based schema
"original_method": {
"name": "Original Method",
"description": "Original baseline extraction method",
"method_type": MethodType.BASELINE,
"schema_type": SchemaType.DIRECT_BASED,
"module_path": "agentgraph.methods.baseline.original_method",
"class_name": "OriginalKnowledgeExtractionMethod",
"supported_features": ["direct_extraction"],
"processing_type": "direct_call"
},
"clustering_method": {
"name": "Clustering Method",
"description": "Clustering-based extraction method",
"method_type": MethodType.BASELINE,
"schema_type": SchemaType.DIRECT_BASED,
"module_path": "agentgraph.methods.baseline.clustering_method",
"class_name": "ClusteringKnowledgeExtractionMethod",
"supported_features": ["direct_extraction", "clustering"],
"processing_type": "direct_call"
},
"direct_llm_method": {
"name": "Direct LLM Method",
"description": "Direct LLM-based extraction method",
"method_type": MethodType.BASELINE,
"schema_type": SchemaType.DIRECT_BASED,
"module_path": "agentgraph.methods.baseline.direct_llm_method",
"class_name": "DirectLLMKnowledgeExtractor",
"supported_features": ["direct_extraction", "llm"],
"processing_type": "direct_call"
},
"hybrid_method": {
"name": "Hybrid Method",
"description": "Hybrid extraction combining multiple approaches",
"method_type": MethodType.BASELINE,
"schema_type": SchemaType.DIRECT_BASED,
"module_path": "agentgraph.methods.baseline.hybrid_method",
"class_name": "HybridKnowledgeExtractionMethod",
"supported_features": ["direct_extraction", "hybrid"],
"processing_type": "direct_call"
},
"pydantic_method": {
"name": "Pydantic Method",
"description": "Pydantic-based extraction method",
"method_type": MethodType.BASELINE,
"schema_type": SchemaType.DIRECT_BASED,
"module_path": "agentgraph.methods.baseline.pydantic_method",
"class_name": "PydanticKnowledgeExtractor",
"supported_features": ["direct_extraction", "pydantic"],
"processing_type": "direct_call"
},
"unified_method": {
"name": "Unified Method",
"description": "Unified extraction method",
"method_type": MethodType.BASELINE,
"schema_type": SchemaType.DIRECT_BASED,
"module_path": "agentgraph.methods.baseline.unified_method",
"class_name": "UnifiedKnowledgeExtractionMethod",
"supported_features": ["direct_extraction", "unified"],
"processing_type": "direct_call"
},
"openai_agent": {
"name": "OpenAI Agent",
"description": "OpenAI Agent with function tools and validation",
"method_type": MethodType.BASELINE,
"schema_type": SchemaType.DIRECT_BASED,
"module_path": "agentgraph.methods.baseline.openai_agent",
"class_name": "OpenAIAgentKnowledgeExtractor",
"supported_features": ["direct_extraction", "pipeline", "validation_improvement", "graph_enhancement"],
"processing_type": "direct_call"
},
"sequential_pydantic": {
"name": "Sequential Pydantic",
"description": "Sequential Pydantic-based extraction method",
"method_type": MethodType.BASELINE,
"schema_type": SchemaType.DIRECT_BASED,
"module_path": "agentgraph.methods.baseline.pydantic_method",
"class_name": "PydanticKnowledgeExtractor",
"supported_features": ["direct_extraction", "pydantic", "sequential"],
"processing_type": "direct_call"
},
"pydantic_hybrid_method": {
"name": "Pydantic Hybrid Method",
"description": "Hybrid Pydantic-based extraction method",
"method_type": MethodType.BASELINE,
"schema_type": SchemaType.DIRECT_BASED,
"module_path": "agentgraph.methods.baseline.pydantic_method",
"class_name": "PydanticKnowledgeExtractor",
"supported_features": ["direct_extraction", "pydantic", "hybrid"],
"processing_type": "direct_call"
},
# rule_based_method removed due to import errors
}
# Default method configuration
DEFAULT_METHOD = "openai_structured"
def get_available_methods() -> Dict[str, Dict[str, Any]]:
"""Get all available methods with their metadata"""
return AVAILABLE_METHODS.copy()
def get_method_info(method_name: str) -> Optional[Dict[str, Any]]:
"""Get information about a specific method"""
return AVAILABLE_METHODS.get(method_name)
def get_methods_by_type(method_type: MethodType) -> Dict[str, Dict[str, Any]]:
"""Get methods filtered by type"""
return {
name: info for name, info in AVAILABLE_METHODS.items()
if info["method_type"] == method_type
}
def get_methods_by_schema(schema_type: SchemaType) -> Dict[str, Dict[str, Any]]:
"""Get methods filtered by schema type"""
return {
name: info for name, info in AVAILABLE_METHODS.items()
if info["schema_type"] == schema_type
}
def get_schema_for_method(method_name: str) -> Optional[SchemaType]:
"""Get the schema type for a specific method"""
method_info = get_method_info(method_name)
return method_info["schema_type"] if method_info else None
def is_valid_method(method_name: str) -> bool:
"""Check if a method name is valid"""
return method_name in AVAILABLE_METHODS
def get_method_names() -> List[str]:
"""Get list of all method names"""
return list(AVAILABLE_METHODS.keys())
def get_production_methods() -> List[str]:
"""Get list of production method names"""
return [
name for name, info in AVAILABLE_METHODS.items()
if info["method_type"] == MethodType.PRODUCTION
]
def get_baseline_methods() -> List[str]:
"""Get list of baseline method names"""
return [
name for name, info in AVAILABLE_METHODS.items()
if info["method_type"] == MethodType.BASELINE
]
def get_method_display_name(method_name: str) -> str:
"""Get display name for a method"""
method_info = get_method_info(method_name)
return method_info["name"] if method_info else method_name
def get_method_description(method_name: str) -> str:
"""Get description for a method"""
method_info = get_method_info(method_name)
return method_info["description"] if method_info else ""
def validate_method_schema_compatibility(method_name: str, expected_schema: SchemaType) -> bool:
"""Validate that a method uses the expected schema type"""
method_schema = get_schema_for_method(method_name)
return method_schema == expected_schema if method_schema else False
|