MLOps using Vertex AI : Beyond Model Training

MLOps using Vertex AI: Beyond Model Training (GDG Mysore Devfest'22)

Transcript

1. Mysuru MLOps using Vertex AI: Beyond Model Training Shadab Hussain Senior Associate - MLOps, TheMathCompany
2. What is Machine Learning?
3. What is Machine Learning? 1. An application of artificial intelligence 2. Built using algorithms and data 3. Automatically analyze and make decision by itself without human intervention.
6. A classification problem is when the output variable is a category. Examples: “red” or “blue”? will it rain today or not? “cat”, “dog” or “tiger”?
7. A regression problem is when the output variable is a real value. Examples: Predict value of a stock? Price of house in a city?
8. Problems with Traditional way for building ML Models • Good configuration hardware required. • Model needs to be deployed in a scalable way. • Machine Learning expertise required to write code • Build efficient models.
10. How to tackle all this??
11. Vertex AI • Train models without code, minimal expertise required • A unified UI for the entire ML workflow • Manage your models with confidence • Pre-trained APIs for vision, video, natural language, and more
12. Vertex AI • Image Classification Object Detection • Tabular Regression/classification Forecasting
13. Vertex AI • Text Classification Entity Extraction Sentiment analysis • Video Classification Action Recognition
14. How does Vertex AI Tables help? • It helps you build and deploy high quality machine learning models on structured data (Tables). • No code required!! • No Machine Learning Expertise required!!
15. Example problem Statement 1. Predicting Housing Prices 2. Predicting possibility of getting Diabetes 3. Credit card data for 'good' or 'bad' customer 4. Mobile Phone Price range from it’s Features (RAM, Battery, etc)
16. https://github.com/techwithshad ab/vertex-ai-wine-demo