You've already forked RekomenciBackend
36 lines
1.4 KiB
Python
36 lines
1.4 KiB
Python
from collections.abc import Sequence
|
|
from typing import override
|
|
|
|
from sqlalchemy import label, select
|
|
from sqlalchemy.ext.asyncio import AsyncSession
|
|
|
|
from template_project.adapters.data_gateways.tables import vacancy_embedding_table, vacancy_table
|
|
from template_project.application.vacancy.data_gateway import VacancyDataGateway
|
|
from template_project.application.vacancy.data_structure import SuitableVacancy
|
|
from template_project.application.vacancy.entity import Vacancy, VacancyEmbedding
|
|
|
|
|
|
class DefaultVacancyDataGateway(VacancyDataGateway):
|
|
def __init__(self, session: AsyncSession) -> None:
|
|
self._session = session
|
|
|
|
@override
|
|
async def get_suitable(self, vector: list[float]) -> Sequence[SuitableVacancy]:
|
|
distance_expr = vacancy_embedding_table.c.vector.cosine_distance(vector)
|
|
similarity_expr = 1 - distance_expr
|
|
statement = (
|
|
select(Vacancy, label("resume_similarity", similarity_expr))
|
|
.join(VacancyEmbedding, vacancy_embedding_table.c.vacancy_id == vacancy_table.c.id)
|
|
.where(similarity_expr >= 0.5)
|
|
.order_by(distance_expr.asc())
|
|
.limit(100)
|
|
)
|
|
result = await self._session.execute(statement)
|
|
return [
|
|
SuitableVacancy(
|
|
vacancy=row[0],
|
|
resume_similarity=row[1],
|
|
)
|
|
for row in result.all()
|
|
]
|