CS231n - Deep Learning for Computer Vision

Stanford - Spring 2023

Assignments

• Assignment 1: Deep Learning Basics
  • Code
    • Q1: k-Nearest Neighbor classifier
    • Q2: Training a Support Vector Machine
    • Q3: Implement a Softmax classifier
    • Q4: Two-Layer Neural Network
    • Q5: Higher Level Representations: Image Features
  • Additional Notes
    • A1: Calculate Distances
    • A2: Support Vector Machine
    • A3: Softmax
    • A4: Neural Network
• Assignment 2: Perceiving and Understanding the Visual World
  • Code
    • Q1: Multi-Layer Fully Connected Neural Networks
    • Q2: Batch Normalization
    • Q3: Dropout
    • Q4: Convolutional Neural Networks
    • Q5: PyTorch on CIFAR-10
  • Additional Notes
    • A1: Neural Network Basics
    • A2: Gray Scale Convolution
    • A3: RGB Scale Convolution
    • A4: Batch RGB Scale Convolution
    • A5: Batch RGB Scale Convolution with Padding and Pooling
    • A6: Convolution Benchmarks
    • A7: Convolution Backward
    • A8: Pooling Layer Details
    • A9: Generate Patches
    • A10: MLP-Mixer
    • A11: MetaFormer
• Assignment3: Generative and Interactive Visual Intelligence
  • Code
    • Q1: Network Visualization: Saliency Maps, Class Visualization, and Fooling Images
    • Q2: Image Captioning with Vanilla RNNs
    • Q3: Image Captioning with Transformers
    • Q4: Generative Adversarial Networks
    • Q5: Self-Supervised Learning for Image Classification
    • Extra Credit: Image Captioning with LSTMs
  • Additional Notes
    • A1: Positional Encoding
    • A2: Attention
    • A3: SimCLR