girafe-ai/ml-course
Ml Course
This repository provides a comprehensive educational framework for mastering machine learning and deep learning through a structured curriculum. It integrates theoretical mathematical foundations—including calculus, probability, and linear algebra—with hands-on laboratory implementations that require learners to build algorithms and neural network architectures from scratch.
The project distinguishes itself by emphasizing first-principles development, ensuring that students understand the underlying mechanics of backpropagation, layer-wise computation, and model optimization. It covers a broad technical scope, ranging from fundamental supervised learning techniques like decision trees and support vector machines to advanced topics such as graph neural networks, sequence modeling, and transformer-based architectures for natural language processing.
Beyond core implementation, the curriculum incorporates systematic data science methodologies, including cross-validation, dimensionality reduction, and performance evaluation techniques. The materials are organized into modular weekly sequences that combine lecture recordings, curated academic literature, and practical programming assignments to facilitate a deep understanding of predictive modeling and model interpretation.
Features
- AI & Machine Learning Education - Provides a structured curriculum of lectures and assignments to master machine learning and deep learning concepts.
- Deep Learning Implementations - Focuses on the manual implementation of neural network architectures from first principles through laboratory exercises.
- Mathematical Foundations - Integrates essential mathematical prerequisites like calculus and linear algebra to support advanced model derivation.
- Machine Learning Curricula - Offers a structured sequence of lectures and programming assignments designed to teach machine learning through hands-on implementation.
- Machine Learning Education - Delivers a structured educational curriculum covering fundamental and advanced machine learning concepts through lectures and assignments.
- Mathematical Foundations - Outlines essential mathematical concepts in calculus, probability, and linear algebra required for machine learning.
- Supervised Learning Tutorials - Teaches essential supervised learning techniques including nearest neighbors and support vector machines through practical implementation.
- Neural Network Architectures - Provides technical documentation and guides for constructing deep learning models and specialized network structures.
- Neural Network Training from Scratch - Implements fundamental neural network architectures and algorithms from scratch to demonstrate underlying mathematical mechanics.
- Bias and Variance Analysis - Teaches systematic performance assessment through bias-variance decomposition and cross-validation procedures.
- Convolutional Neural Networks - Covers the application of convolutional and pooling layers to extract spatial features from image data.
- Decision Trees - Covers the construction of hierarchical decision tree models for classification and regression tasks.
- Dimensionality Reduction Techniques - Covers techniques for projecting high-dimensional datasets into lower-dimensional subspaces to preserve essential structural information.
- End-to-End Training Pipelines - Outlines end-to-end processes for data preparation, model training, and evaluation through structured workflows.
- Ensemble Learning Techniques - Explains techniques for combining multiple predictive models to improve accuracy and reduce variance.
- Graph Neural Network Implementations - Provides foundational resources for applying deep learning techniques to complex graph-structured data.
- Graph Neural Networks - Includes introductory materials and research-based tutorials for applying deep learning techniques to graph-structured data.
- Linear Regression - Provides structured materials and exercises for mastering the foundations of linear regression models.
- Machine Learning Evaluation - Teaches assessment of model performance using cross-validation and feature importance analysis.
- Model Interpretability - Offers tools for evaluating model stability and feature importance to explain decision-making in complex models.
- K-Fold Cross-Validation - Provides methods for evaluating model performance by partitioning data into multiple subsets to ensure robust results.
- Natural Language Processing - Aggregates research and tutorials focused on analyzing and generating human language using modern deep learning architectures.
- Neural Network Regularization - Covers regularization techniques like dropout and batch normalization to improve training stability.
- Recurrent Neural Networks - Provides resources for training recurrent neural networks to predict sequences and generate text.
- Support Vector Machines - Covers margin-based classification techniques and the implementation of support vector machines.
- Deep Learning Resources - Provides comprehensive documentation and implementation guides for building and training neural network architectures.
- Weekly Module Progressions - Structures complex machine learning topics into a progressive sequence of weekly modules and lecture materials.
- Natural Language Processing Tutorials - Provides tutorials and resources for mastering text representation, sequence modeling, and transformer-based architectures.
- Supervised Learning Examples - Features practical exercises and modules for implementing classification, regression, and ensemble algorithms from scratch.
- Laboratory Curricula - Organizes technical exercises into modular, hands-on laboratory tasks that simulate real-world data science workflows.
