Deep Learning & Neural Networks

Perceptron, activation functions and layers

Reading

Feedforward networks and backpropagation

Reading

Loss functions — MSE, cross-entropy, focal

Reading

Optimisers — SGD, Adam, RMSProp

Reading

Learning rate schedules and warm restarts

Reading

Practice — build a neural net from scratch in NumPy

Exercise

Keras Sequential and Functional API

Reading

Regularisation — Dropout, BatchNorm, L2

Reading

Callbacks — EarlyStopping, ModelCheckpoint

Reading

Training on GPU with Colab

Reading

Custom training loops with tf.GradientTape

Reading

Practice — classify tabular data with Keras

Exercise

Convolution, pooling and receptive field

Reading

CNN architectures — VGG, ResNet, EfficientNet

Reading

Transfer learning and fine-tuning

Reading

Data augmentation for images

Reading

Object detection intro — YOLO overview

Reading

Practice — image classifier with transfer learning

Exercise

RNN and vanishing gradient problem

Reading

LSTM and GRU — gating mechanisms

Reading

Time-series forecasting with LSTM

Reading

Seq2Seq and attention mechanism intro

Reading

Temporal Convolutional Networks (TCN)

Reading

Practice — stock price forecasting with LSTM

Exercise

Transformer architecture — self-attention

Reading

Tokenisation and word embeddings

Reading

HuggingFace Transformers library

Reading

Fine-tuning BERT for text classification

Reading

Zero-shot and few-shot learning

Reading

Practice — sentiment analysis with BERT

Exercise

Project brief — image or text classification

Reading

Data pipeline, augmentation and loaders

Exercise

Design, train and evaluate the model

Exercise

Optimise and generate confusion matrix

Exercise

Capstone — deploy model and submit report

Project