Skip to content

Latest commit

 

History

History
19 lines (14 loc) · 1.13 KB

README.md

File metadata and controls

19 lines (14 loc) · 1.13 KB

Amazon SageMaker

For Amazon Sagemaker part, I have initialized the git repo from the terminal of JupyterLab on AWS platform. I added git remote URL to connect JupyterLab terminal to Github server.

Identifying Bees Using Crowd Sourced Data using Amazon SageMaker

Table of contents

Introduction to dataset

  • Labeling with Amazon SageMaker Ground Truth
  • Reviewing labeling results
  • Training an Object Detection model
  • Review of Training Results
  • Model Tuning
  • Cleanup

Need to Know beside following video Getting Started with AWS Machine Learning

  • Make sure to setup your own S3 Bucket
  • Just follow up the course (Getting Started with AWS Machine Learning) and practice on your own. Please do not use demo.ipynb from my repo since I updated it with my information. You need implement your own to learn how python coding works for computer vision.