Data Science and Machine Learning
Data Analysis with Python and Pandas
Analyze data like a pro. Use Python, Pandas, and NumPy to clean, transform, and explore datasets and extract meaningful…
USD 39
Data Science and Machine Learning
Deep Learning with Neural Networks
Build, train and evaluate neural networks in Keras and PyTorch, from dense networks to CNNs, RNNs and transfer learning, and deliver a working image classifier.
About this course
Go deep into AI. Understand neural networks, backpropagation, CNNs, and RNNs, and train models for vision and language tasks.
What you will learn
- Explain how neurons, layers, activation functions, loss functions and backpropagation work together to train a network
- Build and train dense networks in Keras and read the equivalent PyTorch training loop
- Diagnose underfitting and overfitting from learning curves and apply dropout, weight decay and early stopping
- Design convolutional networks for images and recurrent networks for sequences
- Apply transfer learning and data augmentation to reach strong results with small datasets
- Train on GPUs in Google Colab and evaluate models honestly with validation data, test sets and confusion matrices
Curriculum
Module 1Neural Network Foundations4 lessons
- 1From Neuron to Network Free preview
- 2Activation Functions and Layers
- 3How Networks Learn: Backpropagation Intuition
- 4Loss Functions and Optimisers
- QModule quiz (4 questions)
Module 2Building Networks4 lessons
- 5Your First Network in Keras/PyTorch
- 6Training: Batches, Epochs, Learning Rates
- 7Diagnosing Under- and Overfitting
- 8Regularisation: Dropout and More
- QModule quiz (4 questions)
Module 3Specialised Architectures4 lessons
- 9CNNs for Images
- 10RNNs and Sequences
- 11Transfer Learning in Practice
- 12Introduction to Transformers
- QModule quiz (4 questions)
Module 4Applied Deep Learning3 lessons
- 13Working with GPUs and Colab
- 14Data Augmentation
- 15Course Project: Image Classifier
- QModule quiz (4 questions)
PFinal project: Practical Image Classifier with Transfer Learning
Who this course is for
- Learners with basic Python who want a practical, clear introduction to deep learning
- Data analysts and developers moving from classic machine learning to neural networks
- Students and professionals exploring image or text projects for work or research
Requirements
- A computer with internet access and a Google account for Colab, or a local Python setup with TensorFlow or PyTorch
- Comfortable basic Python and some NumPy; school-level algebra is enough for the maths
- Helpful but not required: an introductory machine learning course
What is included
Written lessons
15 lessons, each with a practical exercise and key points to remember.
Module quizzes
A short quiz at the end of each module. Pass mark 70%, and you can retake it.
Final project
Practical Image Classifier with Transfer Learning. A practical brief with steps and a way to check your own work.
Certificate
A certificate of completion from EMFAD Courses with a unique code anyone can verify.
How the certificate works
Complete every lesson and pass every module quiz with 70% or more. Your certificate then appears in your learner area, ready to print or save as PDF. It shows your name, the course title, the completion date and a unique code. It is a certificate of completion from EMFAD Courses, not an accredited qualification.
More in Data Science and Machine Learning
All →Data Science and Machine Learning
Data Science Fundamentals
Enter the world of data. Learn the data science workflow, tools, and thinking needed to turn raw data into valuable insights.
USD 49
Data Science and Machine Learning
Data Visualization with Tableau
Tell stories with data. Build interactive dashboards and charts in Tableau that make complex information clear and…
USD 39