Guide to your own artificial intelligence application in 3 easy steps
summary
Shows how to turn an idea into a working image classifier: crop eyes into four classes (round, monolid/monocular, hooded, almond), train with 200 train / 100 val per class, use transfer learning with VGG16 pretrained on ImageNet and lock convolutional blocks, get roughly 90–95% test accuracy after fine-tuning, and deploy as a web app with Angular + Bootstrap front end and a small Python web server to serve predictions.
AI-written summary · about
chapters (9)
- 0:00Start
- 0:07Deep Learning as an Ai
- 1:14Training the Model on Furniture
- 7:14Eyes Shapes
- 9:25Transfer Learning
- 11:02Convolution Operation
- 15:05Recap
- 18:52Conclusions
- 19:23Demo
related talks
01h 04mBuilding and Training your own Custom Image Recognition AI
01h 02mAI on the Edge
23mBuild a look alike engine with machine learning and Angular
26mMachine Learning and AI for Angular Developers
23mAI is ready for you. Are you ready for AI?
24mDon't want to miss out on the AI wave?
35mHow to AI in JS?
30mUnderstanding What Artificial Intelligence Actually Sees