Skip to content

toelt-llc/COURSE-zhaw-dlcourse-spring2019

Repository files navigation

"Neural Networks and Deep Learning for Life Sciences and Health Applications" course Material

license MIT

(C) 2018 Umberto Michelucci

This github repository contains the course material for the course

Neural Networks and Deep Learning for Life Sciences and Health Applications, An introductory course about theoretical fundamentals, case studies and implementations in python and tensorflow

given in the Spring Semester 2019 at the ZHAW (University of Zürich for applied science).

Google Colab links for github notebooks

Week 1 notebooks

Introduction to Mathematic and Python

Introduction to Matplotlib

Solution to exercises

Derivation of linear regression formula

Week 2 notebooks

Calculus and optimisation

Data analysis example with 3 and 8

Data analysis example

Gradient Descent examples

Week 3

Overloading of operators

Computational Graphs

Exercise solutions

Activation functions and gradient descent

Linear regression with Keras

One dimensional linear regression with TensorFlow

Week 4

Classification with tensorflow

Linear regression with tensorflow

MNIST with logistic regression and tensorflow

Week 5

Overfitting

Zalando dataset

Week 6

Optimizers - Reference

Network Training - Learning rate - Reference

Zalando dataset

Optimizer introduction and exercises

Zalando dataset and decaying learning rate

Batch mini-batch difference

Week 7

Regularization boundary - Reference

Regularization

Small introduction to Keras

Week 8

Metric analysis - Reference

Metric analysis

k-Fold validation

Week 9

Search methods

Tuning with Zalando Dataset

Fundamentals of CNN

Fundamentals of CNNs

Fundamentals of RNN

Fundamentals of RNN

Final Project

Fundamentals of RNN