Applied Deep Learning

Learn how to apply the latest innovative deep learning research and techniques to exciting real-world business problems in Python by designing neural architectures and thereafter training, interpreting and evaluating distributed deep learning models.

Course Details

This course provides a hands-on and in-depth exploration of the industry-standard Python Keras deep learning API built on top of the TensorFlow machine learning library with which to design, build, interpret and evaluate deep learning models applied to real-world business problems and use-cases. This course follows on from our Introduction to Deep Learning course, and enables principal and lead data scientists to apply the mathematical techniques introduced in that course to exciting real-world use-cases including object detection in images and videos, forecasting with time series data, fraud detection and autonomous image generation using generative adversarial networks. This course also details applied techniques for hyperparameter optimisation for deep learning models, and normalisation layers for gradient propagation.

Requirements

• Introduction to Python or equivalent
• Linear Algebra or equivalent
• Statistical Learning or equivalent
• Introduction to Deep Learning or equivalent

Outcomes

• The ability to apply deep learning techniques in Python.
• The ability to design neural architectures and thereafter build, train, interpret and evaluate deep learning models applied to exciting real-world business problems and use-cases including object detection, forecasting, fraud detection and autonomous image generation.
• The ability to fine-tune and optimise hyperparamters for deep learning models.
• Knowledge of the industry-standard Python Keras deep learning API built on top of the TensorFlow machine learning library.

Contact Us