Vertex AI: Qwik Start

 

L2E3: Data Analyst Skills 과정 첫번째 퀘스트이다.

 

Overview

In this lab, you will use BigQuery for data processing and exploratory data analysis and the Vertex AI platform to train and deploy a custom TensorFlow Regressor model to predict customer lifetime value. The goal of the lab is to introduce to Vertex AI through a high value real world use case - predictive CLV. You will start with a local BigQuery and TensorFlow workflow that you may already be familiar with and progress toward training and deploying your model in the cloud with Vertex AI.

Vertex AI is Google Cloud's next generation, unified platform for machine learning development and the successor to AI Platform announced at Google I/O in May 2021. By developing machine learning solutions on Vertex AI, you can leverage the latest ML pre-built components and AutoML to significantly enhance development productivity, the ability to scale your workflow and decision making with your data, and accelerate time to value.

 

이 실습에서는 BigQuery를 데이터 처리 및 탐색적 데이터 분석에 사용하고 Vertex AI 플랫폼을 사용하여 맞춤형 TensorFlow Regressor 모델을 학습 및 배포하여 고객 평생 가치를 예측합니다. 랩의 목표는 높은 가치의 실제 사용 사례인 예측 CLV를 통해 Vertex AI를 도입하는 것입니다. 이미 익숙할 수 있는 로컬 BigQuery 및 TensorFlow 워크플로로 시작하여 Vertex AI를 사용하여 클라우드에서 모델을 학습 및 배포하는 방향으로 진행합니다.

 

Vertex AI는 기계 학습 개발을 위한 Google Cloud의 차세대 통합 플랫폼이며 2021년 5월 Google I/O에서 발표된 AI Platform의 후속 제품입니다. Vertex AI에서 기계 학습 솔루션을 개발하면 최신 ML 사전 구축 구성 요소를 활용할 수 있습니다. AutoML을 사용하면 개발 생산성, 워크플로를 확장하고 데이터를 사용하여 의사 결정을 내리고 가치 실현 시간을 단축할 수 있습니다.

 

Vertex AI 를 활용하여 퀘스트를 진행한다.

AUTOML을 활용하여 빠른 회귀 모델 생성으로 의사결정, 가치 실현을 빠르게 효율적으로 구현한다.

 

Objectives

• Train a TensorFlow model locally in a hosted Vertex Notebook.
• Create a managed Tabular dataset artifact for experiment tracking.
• Containerize your training code with Cloud Build and push it to Google Cloud Artifact Registry.
• Run a Vertex AI custom training job with your custom model container.
• Use Vertex TensorBoard to visualize model performance.
• Deploy your trained model to a Vertex Online Prediction Endpoint for serving predictions.
• Request an online prediction and explanation and see the response.

호스팅된 Vertex Notebook에서 로컬로 TensorFlow 모델을 학습시킵니다. 실험 추적을 위해 관리되는 테이블 형식 데이터 세트 아티팩트를 만듭니다. Cloud Build로 학습 코드를 컨테이너화하고 Google Cloud Artifact Registry로 푸시하세요. 사용자 지정 모델 컨테이너로 Vertex AI 사용자 지정 교육 작업을 실행합니다. Vertex TensorBoard를 사용하여 모델 성능을 시각화합니다. 예측을 제공하기 위해 학습된 모델을 Vertex Online Prediction Endpoint에 배포합니다. 온라인 예측 및 설명을 요청하고 응답을 확인하십시오.

 

gcloud services enable \
  compute.googleapis.com \
  iam.googleapis.com \
  iamcredentials.googleapis.com \
  monitoring.googleapis.com \
  logging.googleapis.com \
  notebooks.googleapis.com \
  aiplatform.googleapis.com \
  bigquery.googleapis.com \
  artifactregistry.googleapis.com \
  cloudbuild.googleapis.com \
  container.googleapis.com

 

VERTAX AI Workbench에서 노트북 생성을 할 수 있다.

 

주피터 랩 터미널 환경에서 작업 가능