Creating and Deploying Machine Learning Models on GCP
- Created By raju2006
- Last Updated December 27th, 2023
- Overview
- Prerequisites
- Audience
- Audience
- Curriculum
Description:
Unlock the power of machine learning across industries with this comprehensive class. Learn to create machine learning models using Scikit-Learn and delve into deep learning with TensorFlow. Discover the art of model saving and version management, and seamlessly deploy your models using Google Cloud Platform's AI Platform for actionable insights and predictions.
Course Code/Duration:
BDT71 / 1 Day
Learning Objectives:
After this course, you will be able to:
- Understand How to Deploy Models on Google Cloud Platform (GCP)
- Understand How Machines Learn
- Understand Structured and Unstructured Data
- Compare AI vs. Machine Learning vs. Deep Learning
- Use Common Machine Learning Algorithms
- Use Scikit-Learn to Create and Train Machine Learning Models
- Use TensorFlow and Keras to Create and Train Deep Learning Models
- Use AI Platform on GCP
- Use Cloud Storage on GCP
- Use AI Platform Notebooks on GCP to Build and Train Scikit-Learn and TensorFlow Machine Learning Models
- Use GCP to Deploy Trained Machine Learning Models
- Understand the fundamental techniques through Demos and hands-on labs
- Python experience
- Basic understanding of Machine Learning
- This course is designed for Software Architects, Developers, Data Engineer, Data Analyst and Machine Learning Engineer.
- This course is designed for Software Architects, Developers, Data Engineer, Data Analyst and Machine Learning Engineer.
Course Outline:
- Course Introduction
- Compare AI vs ML vs DL
- Understanding how machines learn
- Structured vs. Unstructured data
Lab:
- Installing Anaconda and TensorFlow
- Common machine learning algorithms
- Using Scikit-Learn to create and train machine learning models
- .fit()
- score()
- .predict()
Lab:
- Using scikit-learn to build a linear and a logistic regression model
- Saving a Scikit-Learn Model
- pickle (model.pkl)
- joblib (model.joblib)
- Using GCP Cloud Storage to Store Saved Models
Lab:
- Creating a Cloud Storage bucket on GCP and uploading models
- Introducing Keras/TensorFlow
- TensorFlow intro
- Using Keras
Lab:
- Using Keras to builda linear regression and a neural network model
- Saving a TensorFlow Model
- Saving the model in tensorflow format
- Storing model in GCP Cloud Storage
- Create Different Model Versions for Deployment
- Introducing AI Platform
- AI Platform Notebooks
- AI Platform Models
- Deploy Models on GCP
Lab:
- Create an AI Platform model resource and version resource Serve Models from GCP
Lab:
- Create input data and query deployed model for predictions Next steps
Structured Activity/Exercises/Case Studies:
- Installing Anaconda and TensorFlow
- Using scikit-learn to build a linear and a logistic regression model
- Creating a Cloud Storage bucket on GCP and uploading models
- Using Keras to builda linear regression and a neural network model
- Create an AI Platform model resource and version resource
- Create input data and query deployed model for predictions
Training material provided:
Yes (Digital format)
The curriculum is empty
[INSERT_ELEMENTOR id="19900"]