| Name | openvector_dev JSON |
| Version |
0.1.33
JSON |
| download |
| home_page | None |
| Summary | None |
| upload_time | 2025-07-10 10:50:51 |
| maintainer | None |
| docs_url | None |
| author | p00ler |
| requires_python | >=3.11 |
| license | None |
| keywords |
|
| VCS |
|
| bugtrack_url |
|
| requirements |
No requirements were recorded.
|
| Travis-CI |
No Travis.
|
| coveralls test coverage |
No coveralls.
|
# Persona-Memory Subsystem
## Назначение:
Модуль памяти для диалогового ИИ-агента.
Short-term: последние сообщения (RAM, deque).
Long-term: Qdrant + архив (долговременные факты и истории).
## Архитектура:
ShortTermMemory — хранит последние сообщения.
QdrantAdapter — async-слой для поиска/хранения чанков в Qdrant.
MemoryManagerQdrant — бизнес-логика.
EmbeddingProviderSentenceTransformer / Gemini — эмбеддинги.
MemoryService (фасад) — единая точка для верхнего слоя.
Chunk / ChunkPayload — pydantic-модели для хранения.
## Пример рабочего цикла:
```
mem = MemoryService(short_term, memory_manager)
mem.add_short("user", user_msg)
emb = await embedder.get_embedding(user_msg)
long_memories = await mem.get_long(user_id, emb, k=3)
short_ctx = mem.get_short(10)
mem.add_short("gf", answer)
await mem.save_long(user_id, Chunk(
chunk_id=uuid4(), user_id=user_id, chunk_type="type0",
created_at=datetime.utcnow(), last_hit=datetime.utcnow(),
hit_count=0, text=answer, persistent=False
))
```
## Установка и тесты:
```
poetry add ./vector-memory
docker run -d --name qdrant -p 6333:6333 -v qdrant_data:/qdrant/storage qdrant/qdrant
pytest
```
## Конфигурация:
```
QDRANT_HOST, QDRANT_PORT
QDRANT_COLLECTION
VECTOR_SIZE (совпадает с embedding-моделью)
GEMINI_API_KEY (Gemini Embedding-004)
```
## Перед релизом:
- Все тесты проходят (pytest)
- Размеры векторов и коллекции совпадают
- Архивирование/restore, merge, фильтры работают
- Нет deprecated-методов в adapter
## Roadmap:
- Redis-кеш для hit_count
- Курсорный scroll для больших архивов
- gRPC/gateway-адаптер
## Test-Coverage
```
Name Stmts Miss Cover
-----------------------------------------------------
src\bases\memory_manager_abc.py 20 1 95%
src\memory_manager_qdrant.py 54 11 80%
src\memory_manager_ram.py 95 32 66%
src\qdrant_adapter.py 46 8 83%
src\schemas\chunk.py 31 1 97%
src\short_term.py 26 1 96%
-----------------------------------------------------
TOTAL 272 54 80%
```
Raw data
{
"_id": null,
"home_page": null,
"name": "openvector_dev",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.11",
"maintainer_email": null,
"keywords": null,
"author": "p00ler",
"author_email": "liveitspain@gmail.com",
"download_url": "https://files.pythonhosted.org/packages/a1/0a/2807e2a71d27c0f676677f278466756cf7d05fce6a92feb994254152a431/openvector_dev-0.1.33.tar.gz",
"platform": null,
"description": "# Persona-Memory Subsystem\n\n## \u041d\u0430\u0437\u043d\u0430\u0447\u0435\u043d\u0438\u0435:\n\u041c\u043e\u0434\u0443\u043b\u044c \u043f\u0430\u043c\u044f\u0442\u0438 \u0434\u043b\u044f \u0434\u0438\u0430\u043b\u043e\u0433\u043e\u0432\u043e\u0433\u043e \u0418\u0418-\u0430\u0433\u0435\u043d\u0442\u0430.\n\nShort-term: \u043f\u043e\u0441\u043b\u0435\u0434\u043d\u0438\u0435 \u0441\u043e\u043e\u0431\u0449\u0435\u043d\u0438\u044f (RAM, deque).\n\nLong-term: Qdrant + \u0430\u0440\u0445\u0438\u0432 (\u0434\u043e\u043b\u0433\u043e\u0432\u0440\u0435\u043c\u0435\u043d\u043d\u044b\u0435 \u0444\u0430\u043a\u0442\u044b \u0438 \u0438\u0441\u0442\u043e\u0440\u0438\u0438).\n\n## \u0410\u0440\u0445\u0438\u0442\u0435\u043a\u0442\u0443\u0440\u0430:\nShortTermMemory \u2014 \u0445\u0440\u0430\u043d\u0438\u0442 \u043f\u043e\u0441\u043b\u0435\u0434\u043d\u0438\u0435 \u0441\u043e\u043e\u0431\u0449\u0435\u043d\u0438\u044f.\n\nQdrantAdapter \u2014 async-\u0441\u043b\u043e\u0439 \u0434\u043b\u044f \u043f\u043e\u0438\u0441\u043a\u0430/\u0445\u0440\u0430\u043d\u0435\u043d\u0438\u044f \u0447\u0430\u043d\u043a\u043e\u0432 \u0432 Qdrant.\n\nMemoryManagerQdrant \u2014 \u0431\u0438\u0437\u043d\u0435\u0441-\u043b\u043e\u0433\u0438\u043a\u0430.\n\nEmbeddingProviderSentenceTransformer / Gemini \u2014 \u044d\u043c\u0431\u0435\u0434\u0434\u0438\u043d\u0433\u0438.\n\nMemoryService (\u0444\u0430\u0441\u0430\u0434) \u2014 \u0435\u0434\u0438\u043d\u0430\u044f \u0442\u043e\u0447\u043a\u0430 \u0434\u043b\u044f \u0432\u0435\u0440\u0445\u043d\u0435\u0433\u043e \u0441\u043b\u043e\u044f.\n\nChunk / ChunkPayload \u2014 pydantic-\u043c\u043e\u0434\u0435\u043b\u0438 \u0434\u043b\u044f \u0445\u0440\u0430\u043d\u0435\u043d\u0438\u044f.\n\n## \u041f\u0440\u0438\u043c\u0435\u0440 \u0440\u0430\u0431\u043e\u0447\u0435\u0433\u043e \u0446\u0438\u043a\u043b\u0430:\n```\nmem = MemoryService(short_term, memory_manager)\nmem.add_short(\"user\", user_msg)\nemb = await embedder.get_embedding(user_msg)\nlong_memories = await mem.get_long(user_id, emb, k=3)\nshort_ctx = mem.get_short(10)\nmem.add_short(\"gf\", answer)\nawait mem.save_long(user_id, Chunk(\n chunk_id=uuid4(), user_id=user_id, chunk_type=\"type0\",\n created_at=datetime.utcnow(), last_hit=datetime.utcnow(),\n hit_count=0, text=answer, persistent=False\n))\n```\n\n## \u0423\u0441\u0442\u0430\u043d\u043e\u0432\u043a\u0430 \u0438 \u0442\u0435\u0441\u0442\u044b:\n```\npoetry add ./vector-memory\n\ndocker run -d --name qdrant -p 6333:6333 -v qdrant_data:/qdrant/storage qdrant/qdrant\n \npytest\n```\n\n## \u041a\u043e\u043d\u0444\u0438\u0433\u0443\u0440\u0430\u0446\u0438\u044f:\n```\nQDRANT_HOST, QDRANT_PORT\nQDRANT_COLLECTION\nVECTOR_SIZE (\u0441\u043e\u0432\u043f\u0430\u0434\u0430\u0435\u0442 \u0441 embedding-\u043c\u043e\u0434\u0435\u043b\u044c\u044e)\nGEMINI_API_KEY (Gemini Embedding-004)\n```\n\n## \u041f\u0435\u0440\u0435\u0434 \u0440\u0435\u043b\u0438\u0437\u043e\u043c:\n- \u0412\u0441\u0435 \u0442\u0435\u0441\u0442\u044b \u043f\u0440\u043e\u0445\u043e\u0434\u044f\u0442 (pytest)\n- \u0420\u0430\u0437\u043c\u0435\u0440\u044b \u0432\u0435\u043a\u0442\u043e\u0440\u043e\u0432 \u0438 \u043a\u043e\u043b\u043b\u0435\u043a\u0446\u0438\u0438 \u0441\u043e\u0432\u043f\u0430\u0434\u0430\u044e\u0442\n- \u0410\u0440\u0445\u0438\u0432\u0438\u0440\u043e\u0432\u0430\u043d\u0438\u0435/restore, merge, \u0444\u0438\u043b\u044c\u0442\u0440\u044b \u0440\u0430\u0431\u043e\u0442\u0430\u044e\u0442\n- \u041d\u0435\u0442 deprecated-\u043c\u0435\u0442\u043e\u0434\u043e\u0432 \u0432 adapter\n\n## Roadmap:\n- Redis-\u043a\u0435\u0448 \u0434\u043b\u044f hit_count\n- \u041a\u0443\u0440\u0441\u043e\u0440\u043d\u044b\u0439 scroll \u0434\u043b\u044f \u0431\u043e\u043b\u044c\u0448\u0438\u0445 \u0430\u0440\u0445\u0438\u0432\u043e\u0432\n- gRPC/gateway-\u0430\u0434\u0430\u043f\u0442\u0435\u0440\n\n## Test-Coverage\n```\nName Stmts Miss Cover\n-----------------------------------------------------\nsrc\\bases\\memory_manager_abc.py 20 1 95%\nsrc\\memory_manager_qdrant.py 54 11 80%\nsrc\\memory_manager_ram.py 95 32 66%\nsrc\\qdrant_adapter.py 46 8 83%\nsrc\\schemas\\chunk.py 31 1 97%\nsrc\\short_term.py 26 1 96%\n-----------------------------------------------------\nTOTAL 272 54 80%\n```\n\n",
"bugtrack_url": null,
"license": null,
"summary": null,
"version": "0.1.33",
"project_urls": null,
"split_keywords": [],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "116a7a57da3cabd6ee2dd91bcc11c3b2db6aa23ce34e59244586f243987a1186",
"md5": "708d7dc4f0fb6905114766243ef5d615",
"sha256": "4d249e8ca3c6e6dbda4812097541ec6dd6a1dfad44747bdb0f95bf58403b2501"
},
"downloads": -1,
"filename": "openvector_dev-0.1.33-py3-none-any.whl",
"has_sig": false,
"md5_digest": "708d7dc4f0fb6905114766243ef5d615",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.11",
"size": 16684,
"upload_time": "2025-07-10T10:50:49",
"upload_time_iso_8601": "2025-07-10T10:50:49.432050Z",
"url": "https://files.pythonhosted.org/packages/11/6a/7a57da3cabd6ee2dd91bcc11c3b2db6aa23ce34e59244586f243987a1186/openvector_dev-0.1.33-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "a10a2807e2a71d27c0f676677f278466756cf7d05fce6a92feb994254152a431",
"md5": "2fff1c98dc5765a227b278010a356986",
"sha256": "56a62eda307f934b7e25f6ad3abd761d36aa0c55bb6706889041b049ce946859"
},
"downloads": -1,
"filename": "openvector_dev-0.1.33.tar.gz",
"has_sig": false,
"md5_digest": "2fff1c98dc5765a227b278010a356986",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.11",
"size": 12398,
"upload_time": "2025-07-10T10:50:51",
"upload_time_iso_8601": "2025-07-10T10:50:51.017392Z",
"url": "https://files.pythonhosted.org/packages/a1/0a/2807e2a71d27c0f676677f278466756cf7d05fce6a92feb994254152a431/openvector_dev-0.1.33.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-07-10 10:50:51",
"github": false,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"lcname": "openvector_dev"
}
