Short coruse description: Types of data mining: classification, clustering, association, prediction. Processes: data preparation, model building. Techniques: decision tree, neural network, evolutionary computing, Bayesian network. Applications: multi-media, text and web mining.
We will be using Python 3 with TensorFlow.
- Tutorial 01: Introduction to Python
- Tutorial 02 Introduction to NumPy
- Tutorial 2.1 Avoid loops!
- Tutorial 03: Model Selection, Overfitting and Regularization
- Tutorial 04: Fully Connected Neural Networks - synthetic
- Tutorial 05: Fully Connected Neural Networks - MNIST
- Optional:
- Tutorial 06: Softmax, one-hot-encoding, loss functions
- Tutorial 07: Fully Connected Neural Networks - revisited
- Tutorial 08: CNNs - MNIST
- Tutorial 09: CNNs - Gender Classification
- Tutorial 10: Transfer Learning - Garbage Classification
- Tutorial 11: Fully Convolutional Neural Network - Denoising
- Tutorial 12: Medical Image Segmentation - UNET
- Reading week
- Tutorial 13: AutoML
- Tutorial 16: Recurrent Neural Networks - Next Frame Predictions
- Tutorial 17: Generative Adversarial Networks
- Tutorial 18: Generative Adversarial Networks
- Assignment 01 - option 01: Garbage Classification System
- Assignment 01 - option 02: Python/NumPy Programming
- Assignment 02: Paper Reading
- Assignment 03: Coding