Coursera Deep Learning Spezialisation

About

This 5 course specialization is taught by Andrew Ng, the founder of Coursera and one of the pioneers in machine learning. You can notice Andrew Ngs structured mind throughout the course material, which is very focused and well structured. Most algorithms are taught from scratch. Even the the well known Optimizers such as Adam and RMSProb are programmed by students themselves in their exercises.

Course Work

You can find all the work we did in the course as html-jupyther notebooks. It is not always possible to rerun the notebooks themselves though, since I did not always include the data.

Course 1: Neuronal Networks and Deep Learning

• Python Basics with Numpy
• Logistic Regression with a Neural Network mindset
• Planar Data Classification with one Hidden Layer
• Building your Neural Network Step by Step
• Deep Neural Network Application

Course 2: Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization

• Gradient Checking
• Initilization
• Regularization
• Optimization Methods
• Tensorflow Tutorial
• Gradient Checking

Course 3: Structuring Machine Learning Projects

In this course their was no practical work, only some quizzes

• orthogonalization of steps
• setting your goals
• human performance as baseline

Course 4: Convolutional Neural Networks

(not yet done)

Course 5: Sequence Models

(not yet done)