Introduction and reminders about Machine Learning
- Machine Learning and its Applications
- Deep learning and its applications
The basics of TensorFlow
- Tensors
- Variables vs. Placeholders
- Runtime graph
- Session (session, interactiveSession)
- First program in TensorFlow
- Manipulation of data * Visualization of data with Tensorboard
- TensorFlow API: Tf.contrib.learn
- Running on CPUs vs. GPUs
- Running on Cluster
- Going into production with TensorServing
- Lab 1: Manipulation of TensorFlow Basics
Machine Learning with TensorFlow
Regression with TensorFlow
- Use case: Prediction of sale prices of houses
- Linear regression, multiple
- Optimization
- Comparison of models
- Lab 2: Regression
Classification with TensorFlow
- Use case: Image classification - MNIST dataset
- Logistic Regression, Random Forests, ...
- Compare models * Lab 3: Classification
Deep learning
Perceptron and multilayer neural networks
- Motivation
- Use case: Image classification - MNIST dataset
- Principle and operation
- Lab 4: Classification with multilayer networks
Convolutional Neural Networks (CNN)
- Motivation
- Use case: Image classification - MNIST dataset
- Principle and operation
- Lab 5: Image recognition with convolutional networks
Recurrent Neural Networks (RNN)
- Motivation
- Use case: Natural language processing
- Long Short-Term Memory (LSTM)
- Recurrent Neural Networks (RNN)
- Lab 6: Natural Language Processing (NLP) with Recurring Networks
Restricted Boltzmann Machine and Neural Networks Autoencoders
- Motivation
- Use case: Dimension reduction
- Restricted Boltzmann Machine (RBM)
- Deep Belief Network (DBN)
- Lab 7: Dimension reduction with autoencoders