Code repository for the Deep Learning Workshop organized by the DISTRO Innovative Training Network:
Introduction to Deep Learning
Daniele Giunchi,
Carlo Innamorati
Alejandro Sztrajman
Institute of Science and Technology (IST), Vienna, Austria. November 2017.
1) Download Anaconda (https://www.anaconda.com/download/)
2) Install it!
3) Open a terminal as administrator or use sudo
4) Run: conda create -n keras-env python=3
5) Run: activate keras-env
6) Use "pip install <package>" to install the following packages:
tensorflow, keras, pillow, pandas, matplotlib, jupyter, h5py, sklearn
7) Clone this repository: git clone https://github.com/asztr/WienTrainingWeek.git
8) For notebooks, launch "jupyter notebook".
Presentation Slides: Introduction to Convolutional Neural Networks.
Notebook 1: Training a CNN with MNIST.
Notebook 2: RNNs.
Notebook 3: Time Series Prediction with RNNs.
Notebook 4: 2D Convolutional LSTM.
Notebook 5: Sentiment Analysis.
Notebook 6: Visual Question Answering with CNNs.
Notebook 7: Inception LSTM.
Notebook 8: Callbacks.
Notebook 9: Generator Principles.
Notebook 10: Advanced Tensorflow.