Start building deep neural networks using an intuitive Python interface
under development please contact me if you have any questions or feedback
Learn how to use Gluon’s intuitive Python interface to build deep learning models and solve real-world problems. Deep learning is the main driver behind the breakthroughs in Artificial Intelligence, it revolutionized computer vision and natural language processing. The best way to get started with deep learning is by gaining hands-on experience that you can leverage for your own projects.
This course introduces you to deep learning by demonstrating how to implement deep neural networks. This is done by using the Gluon interface. Gluon is a open source deep learning interface that let you build powerful and scalable deep learning models using a simple Python API. The intuitive interface removes the hassle and lets you focus on experimenting and prototyping solutions without compromising performance.
Gluon is developed by Microsoft and Amazon Web Services (AWS) and uses high-level building blocks on top of Apache MXNet and Microsoft Cognitive Toolkit (CNTK).
Popular deep learning architectures, like Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) are discussed throughout the course and practical applications, like image and text classification, are demonstrated. In addition, state-of-the-art deep learning models are discussed and we will show how you can leverage the pre-trained weights of these models. In the last section of this course we will dive deeper into the world of computer vision by demonstrating how to use Gluon when implementing deep learning models for facial recognition, object detection, and semantic segmentation.
- Introduction & Getting Started
- Build your first neural network with Gluon
- Convolutional Neural Networks (CNNs)
- Recurrent Neural Networks (RNNs)
- Deep Reinforcement Learning (RL)
- Generative Adversarial Networks (GANs)
- Deep learning for Computer Vision with Gluon
- Optimize, deploy, and scale deep learning models (in the cloud)
- Bringing deep learning models into production
- Advanced deep learning topics & applications
Everyone has their own learning style. Therefore, we try to provide a diverse set of learning materials so that everyone can pick the right (combination of) materials that works for them. The basis of the materials are the Jupyter Notebooks. Jupyter Notebooks are a great tool to explore deep learning and machine learning in general. The interactive feedback provided within these notebooks will allow users to iterate and experiment fast in an user friendly environment - on local machines and cloud instances.
- Jupyter Notebooks
- Python code files
- Walk-through videos
- Book (printed and ebook)
- Regular Q&A sessions (if there is enough interest)
- Meetups (later stage)
- Python
- Machine learning
- Mathematics (calculus and linear algebra)
For successful completion of this course, students will require the computer systems with at least the following:
- 64-bit operating system
- RAM 8GB+
For an optimal experience with hands-on labs and other practical activities, we recommend the following configuration:
- 64-bit operating system
- RAM 16GB+
- NVIDIA graphics card (CUDA-Enabled) or access to cloud solution
- Python 3
- Jupyter Notebook
- GitHub: @indradenbakker
- Twitter: @indradenbakker
- Kaggle: @indradenbakker
- Medium
- Amazon