Highschool_ML_Course

A Machine Learning course specifically for highschool students based on the USAAIO course provided by Beaver-Edge AI.

IMPORTANT:If you want best experience please download the whole repo and run the ipynb for visualization!!!

---

Table Of Content

• 0. Prerequisites
  • 0.1 Environment Setup (Anaconda, CUDA, VS Code, Python)
• 1. Mathematical Foundations for AI
  • 1.1 Linear Algebra
    • 1.1.1 Vector Spaces, Subspaces, Basis, Orthonormal Vectors
    • 1.1.2 Vector and Matrix Operations
    • 1.1.3 Eigenvalues and Eigenvectors
    • 1.1.4 Matrix Decompositions
  • 1.2 Calculus
    • 1.2.1 Single-Variable Derivatives
    • 1.2.2 Multivariable Derivatives and Gradients
    • 1.2.3 Chain Rule
  • 1.3 Probability & Statistics
    • 1.3.1 Discrete Distributions
    • 1.3.2 Continuous Distributions
    • 1.3.3 Expectation and Mean
    • 1.3.4 Variance and Covariance
    • 1.3.5 Bayes’ Rule
  • 1.4 Convex Optimization
    • 1.4.1 Convexity
    • 1.4.2 Gradient Descent
    • 1.4.3 Duality
• 2. Python for AI
  • 2.1 Advanced Python Techniques
  • 2.2 NumPy
  • 2.3 Pandas
  • 2.4 Matplotlib
  • 2.5 Seaborn
• 3. Core Machine Learning
  • 3.1 Terminology (Supervised, Unsupervised, Overfitting, etc.)
  • 3.2 Linear Regression
  • 3.3 Logistic Regression
  • 3.4 Regularization, Bias–Variance Trade-off, Kernel Methods
  • 3.5 Cross-Validation
  • 3.6 k-Nearest Neighbors
  • 3.7 K-Means Clustering
  • 3.8 Support Vector Machines
  • 3.9 Principal Component Analysis & Dimensionality Reduction (done)
  • 3.10 Decision Trees
  • 3.11 Random Forests
  • 3.12 Boosting
• 4. PyTorch Fundamentals
  • 4.1 Tensors
  • 4.2 Autograd
  • 4.3 Devices (CPU/GPU)
  • 4.4 Modules
  • 4.5 Datasets
  • 4.6 DataLoader & Collation
  • 4.7 Loss Functions
  • 4.8 Optimizers
• 5. Deep Learning & Computer Vision
  • 5.1 Multi-Layer Perceptron (MLP)
  • 5.2 Forward Propagation & Activation Functions
  • 5.3 Backpropagation & Gradient Descent (done)
  • 5.4 Adam & Other Adaptive Optimizers
  • 5.5 Parameter Initialization
  • 5.6 Batch Normalization
  • 5.7 Dropout
  • 5.8 Convolutional Layers & Pooling Layers (done)
  • 5.9 Convolutional Neural Networks (CNNs) (done)
  • 5.10 Image Data Augmentation
  • 5.11 VGG
  • 5.12 ResNet
  • 5.13 GoogLeNet (Inception)
  • 5.14 Transfer Learning
• 6. Transformers
  • 6.1 Self-Attention
  • 6.2 Cross-Attention
  • 6.3 Masked Self-Attention
  • 6.4 Layer Normalization
  • 6.5 Word Embeddings
  • 6.6 Positional Encoding
  • 6.7 Batch Processing
  • 6.8 Training Procedures
  • 6.9 Inference & Deployment
  • 6.10 Pre-training
  • 6.11 Fine-tuning (Hugging Face)
  • 6.12 BERT, T5, GPT
• 7. Natural Language Processing & Graph Neural Networks
  • 7.1 Character Tokenization
  • 7.2 Subword Tokenization
  • 7.3 Word Tokenization
  • 7.4 Word Embedding Methods
    • 7.4.1 Skip-Gram
    • 7.4.2 Continuous Bag-of-Words
    • 7.4.3 Global Vectors (GloVe)
  • 7.5 Encoder-Only Transformers (BERT)
  • 7.6 Decoder-Only Transformers (GPT)
  • 7.7 Message-Passing Neural Networks
  • 7.8 Graph Convolutional Networks
  • 7.9 Vision Transformers
• 8. OpenCV & Generative AI
  • 8.1 Object Detection
  • 8.2 Adversarial Attacks
  • 8.3 U-Net
  • 8.4 Autoencoders
  • 8.5 Variational Autoencoders
  • 8.6 Generative Adversarial Networks
  • 8.7 Denoising Diffusion Probabilistic Models
  • 8.8 Stable Diffusion