Scalability API Reference¶
Overview¶
The opifex.scalability package provides components for scaling scientific machine learning workflows, including distributed computing, load balancing, orchestration, and a neural functional search engine.
Search Engine¶
The search engine provides full search capabilities for the neural functional registry, including text search, semantic search, filtering, and recommendation systems.
Search Engine for Neural Functional Discovery.
Provides full search capabilities for the neural functional registry including text search, semantic search, filtering, and recommendation systems.
dataclass
¶
SearchQuery(query_text: str = '', functional_type: str | None = None, domain: str | None = None, tags: list[str] | None = None, author_id: str | None = None, min_rating: float | None = None, min_accuracy: float | None = None, max_memory_gb: int | None = None, gpu_required: bool | None = None, limit: int = 50, offset: int = 0, search_type: SearchType = HYBRID)
Structured search query for neural functionals.
dataclass
¶
SearchResult(functional_id: str, name: str, description: str, functional_type: str, author_id: str, tags: list[str], relevance_score: float, metadata: dict[str, Any])
Search result with relevance scoring.
SearchEngine(registry_service: Any, enable_semantic_search: bool = True, similarity_threshold: float = 0.7)
Neural functional search engine with semantic capabilities.
Provides text search, semantic search, filtering, and recommendation systems for discovering neural functionals in the registry.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
registry_service
|
Any
|
Registry service for data access |
required |
enable_semantic_search
|
bool
|
Whether to enable semantic search |
True
|
similarity_threshold
|
float
|
Minimum similarity for semantic matches |
0.7
|
async
¶
search(query: SearchQuery) -> list[SearchResult]
Execute search query and return ranked results.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
query
|
SearchQuery
|
Structured search query |
required |
Returns:
| Type | Description |
|---|---|
list[SearchResult]
|
List of search results sorted by relevance |
async
¶
suggest_functionals(functional_id: str, limit: int = 10) -> list[SearchResult]
Suggest similar functionals based on a given functional.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
functional_id
|
str
|
ID of reference functional |
required |
limit
|
int
|
Maximum suggestions to return |
10
|
Returns:
| Type | Description |
|---|---|
list[SearchResult]
|
List of similar functionals |
async
¶
search_by_problem(problem_description: str, domain: str | None = None, limit: int = 20) -> list[SearchResult]
Search functionals suitable for a specific problem.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
problem_description
|
str
|
Natural language problem description |
required |
domain
|
str | None
|
Scientific domain (optional) |
None
|
limit
|
int
|
Maximum results to return |
20
|
Returns:
| Type | Description |
|---|---|
list[SearchResult]
|
List of functionals suitable for the problem |