Bang Ti Lenydoor

Deep Learning & Neural Networks Professional Certificate

Wishlist Share
Deep Learning & Neural Networks course cover

About Course

Deep Learning & Neural Networks takes you from the basic idea of an artificial neuron to the architectures behind modern AI: convolutional networks that recognise images, recurrent networks that process sequences, and transformers that power large language models such as ChatGPT and Claude.

You will build real models in Python with TensorFlow and Keras in Google Colab, using free GPUs in your browser, so no special hardware is needed. You will learn not only how to build networks, but how to train them well: choosing optimisers and learning rates, preventing overfitting with dropout and data augmentation, using batch normalisation, debugging poor results and reusing powerful pretrained models through transfer learning.

The course uses examples and datasets relevant to Africa, and its capstone uses the Cassava Leaf Disease dataset collected in Uganda by the Makerere University AI Lab, one of Africa’s best-known contributions to applied deep learning.

This course is for learners who have completed our Machine Learning Fundamentals course or have equivalent experience with Python, scikit-learn and model evaluation. How the course works: six modules, 23 lessons with more than 40 videos from leading educators including 3Blue1Brown, StatQuest and DeepLearning.AI, hands-on Colab notebooks, a quiz after every module, a capstone project and a final certification examination. Most learners finish in 10 to 12 weeks at 4 to 5 hours per week.

Show More

What Will You Learn?

  • Explain how artificial neurons, layers and activation functions work
  • Describe how networks learn through loss functions, gradient descent and backpropagation
  • Build, train and evaluate neural networks in TensorFlow and Keras
  • Choose optimisers and learning rates and diagnose training problems from loss curves
  • Prevent overfitting with dropout, early stopping, regularisation and data augmentation
  • Build convolutional neural networks for image classification
  • Apply transfer learning and fine-tuning with pretrained models such as MobileNet
  • Explain recurrent networks, LSTMs, word embeddings and attention
  • Describe how transformers and large language models work
  • Save, convert and deploy models for web and mobile use, responsibly

Course Content

Module 1: Foundations of Neural Networks

  • From Machine Learning to Deep Learning
  • Neurons, Layers and Activation Functions
  • How Networks Learn: Loss, Gradient Descent and Backpropagation
  • Your First Neural Network in Keras
  • Module 1 Quiz: Foundations of Neural Networks

Module 2: Training Deep Networks Well

Module 3: Convolutional Neural Networks for Images

Module 4: Sequence Models

Module 5: Transformers, Deployment and Responsible Deep Learning

Module 6: Capstone Project and Certification

Scroll to Top