Probably the most exhaustive and updated Deep Vision Program in the world! It is spread over three semester-style phases, each restricted by a qualifying exam from https://theschoolof.ai/
- Background & Basics: Machine Learning Intuition
- Python: Python 101 for Machine Learning
- DNN Concepts: Convolutions, Pooling Operations & Channels
- PyTorch: PyTorch 101 for Vision Machine Learning
- First Neural Network: Kernels, Activations, and Layers
- Architectural Basics: We go through 9 model iterations together, step-by-step to find the final architecture
- BN, Kernels & Regularization: Mathematics behind Batch Normalization, Kernel Initialization, and Regularization
- Advance Convolutions, Attention and Image Augmentation: Depthwise, Pixel Shuffle, Dilated, Transpose, Channel Attention and Albumentations Library
- Advanced Training Concepts: Class Activation Maps, Optimizers, LR Schedules, LR Finder & One Cycle Policy
- ResNets: Training ResNet for TinyImageNet from scratch
- Object Detection YoloV2/V3/V4: Understanding YOLO Loss Function & Training Yolo
- The Dawn Of Transformers: Convolutions, Transformers and Types of Attention (Soft, Spatial, Channel, Self and Multi-head)
- Hands-On: Transformers and Attention Mechanism
- Hands-On: Vision Transformers (ViT)
- Modern Object Detection: End-To-End Object Detection with Transformers
- CapStone: Qualifying Project for Phase 2