The subjects in these lists do not represent the order in which the subject will be taught. Some of the subjects are marked as optional, which will be taught if time permits.
* Mathematical Background
Linear Algebra
Probability
Common distributions
Introduction to optimization
Estimation theory
Probabilities measures
* Scientific Python & Programming Tools
Python: OOP, Packages & Modules, Contexts, Iterators, etc.
Scientific Python Eco System: NumPy, SciPy, Pandas, SciKit Learn, SciKit Image, PyTorch, etc.
Local Workspace / Environment
IDE: VS Code, Jupyter, Google Collaboratory.
Working with GPUs
* Classical Machine Learning: Supervised Learning
Concepts:
Algorithms:
Ensemble Methods: Boosting and Bagging.
Classification Score: confusion matrix, precision & recall, F1, AUC, ROC, Precision Recall Curve.
* Classical Machine Learning: Unsupervised Learning
Non Parametric Density Estimation
Parametric Clustering
Non Parametric Clustering
Principal Component Analysis
Manifold Learning (Non Linear Dimensionality Reduction)
Anomaly Detection:
Explainability
* Methods in Neural Networks: Part I
Deep Learning
PyTorch
Feed Forward Networks
Optimization methods
Optimization Scheduling
Regularization Methods
* Methods in Neural Networks: Part II (Computer Vision)
Convolutional NN
Normalization Layers
CNN Architectures
Hands On Tips
Transfer Learning
Supervision Concepts
Large Vision Models and the Vision Transformer (ViT).
Explainability
Production
* Computer Vision Workshop
Self Supervised Learning with Auto Encoders
Object Detection
Image Segmentation
Object Tracking with Large Vision Models (Optional).
Contrastive Learning for Object Recognition (Optional).
* Natural Language Processing (NLP) Workshop
Intro to NLP
Recursive Neural Networks (RNN)
Transformers and Attention.
Working with Embeddings.
Textual Visual Search with CLIP.
