스파클링소다 4.0 Rest App

작성자 박형춘 수정일 2024-02-16 15:34

들어가며

이전 아티클에서 모델 저장소인 model assets에 등록된 모델을 실제로 서비스하는 방법을 알아보았습니다.
이번 아티클에서는 모델서비스를 활욯하기 위해 restapp을 구성하는 방법을 알아보겠습니다.

포탈 접속은 고객사에서 사전 신청한 도메인(url)로 접속합니다.초기 포탈 로그인 정보는 다음과 같습니다.
User ID : superuser
Password : superuser$01

지난 아티클에서 관리계정 또는 일반 계정을 생성하였다면 해당 계정으로 접속합니다.

Restapp

Rest App은 모델서비스 앞단에 위치하여 외부 Client의 rest 요청을 받아 모델서비스로 연결합니다.
모델서비스를 활용하기 위해 Custom 가능한 어플리케이션으로 data 전처리/후처리 코드를 작성합니다.
아키텍처는 아래와 같습니다.

Rest 프로젝트 생성

Rest 서비스 코드 작성을 위한 프로젝트를 생성합니다.
Rest 프로젝트의 정보를 입력하고 저장합니다.
A. General : Rest 프로젝트의 기본 정보를 입력합니다.

- Rest App Name : Rest App의 이름을 지정합니다.

- Image : Rest App이 실행되는 컨테이너의 도커이미지를 선택합니다.

- Description : 자세한 설명을 입력합니다.

B. Resource : Rest App이 사용할 리소스를 지정합니다.

일반적으로 Rest app은 gpu 자원이 필요하지 않으므로 cpu 노드를 선택합니다.

C. Volume : Rest App 컨테이너와 호스트 서버 볼륨을 마운트합니다.

호스트 서버에 존재하는 파일을 Restapp 컨테이너와 공유하거나 Restapp 내부에 생성되는 Output 파일을 호스트 서버와 공유하기 위해 볼륨을 마운트할 수 있습니다.

설정을 마치고 저장하면 restapp 컨테이너가 생성되며 정상적으로 Running 상태가되면 'Running'을 클릭하여 노트북 환경으로 이동합니다.

모델 서비스 설정

Restapp 노트북에서 실행중인 model service와 연결하기 위한 설정을 진행합니다.
app/configs/app_config.yaml 의 내용을 수정합니다.
- mode : 기본 값인 test를 사용합니다.

- log_level : restapp 실행시 출력할 로그 레벨을 지정합니다.

TRACE > DEBUG > INFO > WARN > ERROR > FATAL

INFO로 셋팅하면, INFO, WARN, ERROR, FATAL은 출력됩니다.

FATAL : 아주 심각한 에러가 발생한 상태를 의미합니다.
ERROR : 어떠한 요청을 처리하는 중 문제가 발생하여 즉시 조치가 필요한 상태를 의미합니다.
WARN : 프로그램의 실행에는 문제가 없지만, 향후 시스템 에러를 유발할 가능성이 있는 상태를 의미합니다.
INFO : 어떠한 상태 변경과 같은 정보성 메시지를 의미합니다.
DEBUG : 개발시 디버그 용도로 사용하는 메시지를 의미합니다.
TRACE : 디버그 레벨이 너무 광범위한 것을 해결하기 위해서 좀 더 상세한 이벤트를 출력합니다.

- model_service : 이전 아티클에서 생성한 모델서비스의 이름을 입력합니다.

이 필드에 지정한 모델서비스를 restapp 생성시 호출합니다.

- model_name : 모델서비스에서 사용하는 모델의 이름을 입력합니다.

- model_version : 모델서비스에서 사용하는 모델의 버전을 입력합니다.

모델 서비스 Input/Output 코드 작성

연결될 모델서비스의 인터페이스를 구성합니다.
app/mod_service/services.py의 기본 코드를 지우고 코드를 작성합니다.

아래의 예시 코드를 작성합니다.

import numpy as np
from app.config import cfg

# SoDAFlow Model Servce
from app.common.model_service import SodaModelService

def predict(param=None):
    # configs/app_config.yaml - model connection
    model_service = SodaModelService(
        server_url=cfg.model_service,
        model_name=cfg.model_name,
        model_version=cfg.model_version
    )

    # input setting
    input_data = np.reshape(eval(param), (1,1,28,28)).astype(np.float32)

    # model input, output setting
    model_service.add_input(name='0', shape=(1,1,28,28), dtype='FP32', tensor=input_data)
    model_service.add_output(name='27')

    # inference
    results = model_service.predict()

    return { 'predict': results }

REST API 및 Data 전처리/후처리 코드 작성

data 전처리/후처리 코드를 작성합니다.
app/mod_service/controller.py에 기본 코드를 지우고 코드를 작성합니다
아래의 예시 코드를 작성합니다. (예시 코드에서는 전처리 코드를 생략합니다.

import numpy as np
import json

from flask import Blueprint, request, jsonify
from app.mod_service import services
from app.common.utils import as_json

# Define the blueprint: 'train', set its url prefix: app.url/test
mod_service = Blueprint('service', __name__, url_prefix='/api')

# Set the route and accepted methods
@mod_service.route('/predict/', methods=['GET', 'POST'])
@as_json
def predict():
    try:
        output = None

        if request.method == 'GET':
            if len(request.args.to_dict()) == 0:
                return "no parameter(get)"
            else:
                # get type parameter
                param = request.args["image_data"] 
        else:
            param = request.get_json()
            
            if param == None:
                return "no parameter(post)"
            
            else:
                # post type(json) parameter
                param = param['image_data']               

        # call inference(services.py/predict)
        output = services.predict(param)

        # data post-processing.
        post_data = output['predict'][0]['27']
        # json to array
        result = json.loads(post_data)
        
        # return
        return np.argmax(result)

    except Exception:
        result = dict()
        result['success'] = False
        result['log'] = traceback.format_exc()

        return result

REST API 코드 점검

작성한 restapp 코드가 정상적으로 동작하는지 테스트합니다.
test_client.sh에 REST API 호출 url을 작성합니다.

아래의 코드를 추가합니다.

# label 9
#curl -X GET http://localhost:8088/api/predict/?'image_data'='0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,38,145,195,255,254,203,45,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,17,164,243,253,236,220,253,253,245,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,28,199,254,223,96,29,13,79,249,253,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,17,224,253,195,40,0,0,0,0,207,253,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,198,254,203,34,0,0,0,0,0,208,254,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,81,251,248,56,0,0,0,0,0,43,249,253,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,237,253,113,0,0,0,0,0,0,55,253,244,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,254,253,46,0,0,0,0,34,119,222,253,128,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,170,254,245,95,162,161,195,229,254,254,239,25,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,7,216,253,253,254,253,253,236,254,253,213,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,25,129,230,230,145,96,38,254,253,96,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,93,254,219,13,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,185,254,207,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,106,247,241,102,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,157,253,184,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,68,240,253,84,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,164,254,195,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,114,251,248,64,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,203,253,230,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,186,253,162,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0'

# label 4
curl -X POST http://localhost:8088/api/predict/ -H "Content-Type: application/json" \
    -d '{"image_data":"0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,22,192,134,32,0,0,0,0,0,0,0,0,15,77,5,0,0,0,0,0,0,0,0,0,0,0,0,17,235,250,169,0,0,0,0,0,0,0,0,15,220,241,37,0,0,0,0,0,0,0,0,0,0,0,20,189,253,147,0,0,0,0,0,0,0,0,0,139,253,100,0,0,0,0,0,0,0,0,0,0,0,0,70,253,253,21,0,0,0,0,0,0,0,0,43,254,173,13,0,0,0,0,0,0,0,0,0,0,0,22,153,253,96,0,0,0,0,0,0,0,0,43,231,254,92,0,0,0,0,0,0,0,0,0,0,0,0,163,255,204,11,0,0,0,0,0,0,0,0,104,254,158,0,0,0,0,0,0,0,0,0,0,0,0,0,162,253,178,5,0,0,0,0,0,0,9,131,237,253,0,0,0,0,0,0,0,0,0,0,0,0,0,0,162,253,253,191,175,70,70,70,70,133,197,253,253,169,0,0,0,0,0,0,0,0,0,0,0,0,0,0,51,228,253,253,254,253,253,253,253,254,253,253,219,35,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,17,65,137,254,232,137,137,137,44,253,253,161,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,34,254,206,21,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,160,253,69,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,85,254,241,50,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,158,254,165,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,231,244,50,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,104,254,232,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,208,253,157,0,13,30,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,208,253,154,91,204,161,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,208,253,254,253,154,29,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,61,190,128,23,6,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0"}'

작성한 rest api 호출을 하기 위해 local_rest_server를 실행합니다.

python test_run.py

터미널은 Ctrl + Shift + ` 단축키로 열수 있습니다.

새로운 터미널을 열고 test_client.sh 스크립트를 실행합니다.
```
sh test_client.sh
```
모델서비스에 요청 후 결과 값인 '4'를 정상적으로 반환하였으며 local_server 로그에도 http 상태코드 '200'으로 정상 처리 되었음을 확인할 수 있습니다.

REST API 코드 gitlab에 저장

노트북에서 테스트한 코드를 restapp을 실행할 때 적용하기 위해 gitlab에 저장합니다.

git add . 
git commit -m 'ned-test-app'
git push

REST 서비스 실행

실제 REST 서비스를 실행합니다.

마무리

마무리 내용