Skip to main content

Inference utilities for keynet - Triton Inference Server integration

Project description

keynet-inference

Triton Inference Server 통합과 보안 함수 실행을 위한 추론 유틸리티

설치

pip install keynet-inference

주요 기능

🔌 MLflow-Triton 통합

  • MLflow 모델을 Triton Inference Server로 자동 배포
  • Python 함수를 Triton 모델로 변환
  • S3/MinIO 기반 모델 저장소 지원

🛡️ 보안 함수 실행

  • 격리된 가상환경에서 안전한 실행
  • 메모리 및 CPU 사용량 제한
  • 타임아웃 기반 실행 제어

🛠️ Keynet CLI

  • 함수 검증 및 테스트
  • 배포 자동화
  • 인증 관리
  • Python 3.9~3.12 지원

사용 예제

Python 함수 작성

from keynet_inference.function import keynet_function

@keynet_function(
    python_version="3.11",
    requirements=["numpy", "pandas"],
    timeout=30
)
def process_data(args):
    """데이터 처리 함수"""
    data = args.get("data", [])
    result = sum(data) / len(data) if data else 0

    return {
        "result": result,
        "count": len(data)
    }

CLI 사용

# 함수 검증
keynet validate my_function.py

# 테스트 실행
keynet test my_function.py --params '{"data": [1, 2, 3, 4, 5]}'

# 배포
keynet deploy my_function.py --name my_model

# 인증 관리
keynet login https://api.example.com
keynet logout --all

MLflow 플러그인 사용

import mlflow
from keynet_inference import TritonPlugin

# Triton으로 모델 배포
mlflow.deployments.create_deployment(
    name="my-deployment",
    model_uri="models:/my_model/1",
    flavor="triton",
    config={
        "triton_url": "localhost:8001",
        "model_repository": "s3://models"
    }
)

API 문서

자세한 API 문서는 GitHub Wiki 참조

라이선스

MIT License

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

keynet_inference-0.0.2.dev0.tar.gz (52.1 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

keynet_inference-0.0.2.dev0-py3-none-any.whl (32.7 kB view details)

Uploaded Python 3

File details

Details for the file keynet_inference-0.0.2.dev0.tar.gz.

File metadata

  • Download URL: keynet_inference-0.0.2.dev0.tar.gz
  • Upload date:
  • Size: 52.1 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: python-httpx/0.28.1

File hashes

Hashes for keynet_inference-0.0.2.dev0.tar.gz
Algorithm Hash digest
SHA256 ff08afb5a3f87921c803705f04234bd9fb8cd944ea833e9573c231edba5d8e7f
MD5 9b5683a1b7e8568f6b99710005336aa1
BLAKE2b-256 f5d5e6dea0a13f030d753eea2b1ead9c8e730642cb95e5cc6f9fadbc572e820b

See more details on using hashes here.

File details

Details for the file keynet_inference-0.0.2.dev0-py3-none-any.whl.

File metadata

File hashes

Hashes for keynet_inference-0.0.2.dev0-py3-none-any.whl
Algorithm Hash digest
SHA256 4d40c45685954f450c7f9fa3fbcb6e41a6cabba9ed8c9a836dc8fdb42a13e884
MD5 c22f5058e2cbce775fdf92f2229e514f
BLAKE2b-256 f382f6f2ad484d3ea338a64770d1bbed41aa95d43f1812b7273ce96636a9a8f7

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page