AWS Certified Generative AI Developer – Professional Exam Prep and Study Guide | Course Playlist - Image Feature Extraction with Pre-Trained CNN Models | Local Python + Amazon SageMaker Tutorial

https://github.com/knodax-labs-demo/aws-data-and-ml-labs/blob/main/18-image-feature-extraction-using-pretrained-cnn-in-sagemaker-studio-lab.md Learn how to extract high-quality image features using a pre-trained CNN model (ResNet50) both locally and on Amazon SageMaker. This hands-on tutorial guides you through the entire workflow—from loading a pre-trained model, preprocessing images, extracting feature vectors, and exporting them for downstream machine learning tasks. 🔹 Part 1 — Local Feature Extraction (Python / IntelliJ / Jupyter Notebook) Set up a Python environment Create a Jupyter Notebook Install and import required libraries Load the ResNet50 pre-trained CNN model Preprocess images (resize, normalize, batch) Extract image feature vectors Save features to a CSV file for ML tasks such as classification, clustering, similarity search, and visualization 🔹 Part 2 — Feature Extraction on Amazon SageMaker Create a SageMaker Domain Launch SageMaker Studio Start a Jupyter Notebook on an ml.t3.medium instance Upload images into Studio Run the same preprocessing and feature extraction workflow Save extracted features Clean up SageMaker resources to avoid costs ✨ What You Will Learn How pre-trained CNN models like ResNet50 work Why transfer learning is powerful for ML projects How image preprocessing affects model results How to extract structured ML-ready features from raw images How to run ML workflows inside SageMaker How to delete AWS resources safely 🔧 Tools & Technologies Used Python Jupyter Notebook Amazon SageMaker AWS CLI If you’re preparing for AWS Machine Learning, Data Engineer, or Developer Associate exams, this tutorial gives you practical experience with both deep learning and AWS ML workflows. 👍 Like, subscribe, and leave a comment if you want more step-by-step AWS + machine learning tutorials!