Facebook DETR | ML Coding Series | End to end object detection with transformers
❤️ Become The AI Epiphany Patreon ❤️ https://www.patreon.com/theaiepiphany 👨👩👧👦 Join our Discord community 👨👩👧👦 https://discord.gg/peBrCpheKE The second video in the ML coding series! The longest video I ever made! In this video I do a code walkthrough of Facebook's DETR model from the "End-to-End Object Detection with Transformers" paper. Let me know what you'd like me to cover next! ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ ✅ DETR GitHub: https://github.com/facebookresearch/detr ✅ DETR Paper: https://arxiv.org/abs/2005.12872 ✅ Mini COCO: https://github.com/giddyyupp/coco-minitrain I edited the download/sampling scripts and ended up downloading a valid dataset that has only 100 images. ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ ⌚️ Timetable: 00:00:00 DETR model recap 00:07:30 DETR demo notebook 00:18:55 Visualizing attention notebook 00:29:50 Visualizing encoder attention 00:36:45 Going through the training script 00:38:45 Backbone construction 00:42:05 DETR construction 00:54:00 Data loading and nested tensors 01:02:30 Forward pass through ResNet backbone 01:09:00 Forward pass through the transformer 01:22:00 Hungarian matching algorithm 01:32:50 Loss calculation 01:40:50 Outro ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ 💰 BECOME A PATREON OF THE AI EPIPHANY ❤️ If these videos, GitHub projects, and blogs help you, consider helping me out by supporting me on Patreon! The AI Epiphany - https://www.patreon.com/theaiepiphany One-time donation - https://www.paypal.com/paypalme/theaiepiphany Huge thank you to these AI Epiphany patreons: Eli Mahler Petar Veličković ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ 💼 LinkedIn - https://www.linkedin.com/in/aleksagordic/ 🐦 Twitter - https://twitter.com/gordic_aleksa 👨👩👧👦 Discord - https://discord.gg/peBrCpheKE 📺 YouTube - https://www.youtube.com/c/TheAIEpiphany/ 📚 Medium - https://gordicaleksa.medium.com/ 💻 GitHub - https://github.com/gordicaleksa 📢 AI Newsletter - https://aiepiphany.substack.com/ ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ #detr #transformers #coding
