Practical Introduction To The World Of Computer Vision And Deep Learning With TensorFlow & Keras
Published by O'Reilly Media, Inc.
Implementing neural networks and machine learning techniques
Computer vision has existed since the 1960s and has since evolved into one of the prevalent fields in machine learning and artificial intelligence. Today, solving classical computer vision problems such as face detection, pose estimation, object detection, and semantic segmentation have become trivial thanks to the advancement of deep learning.
Join CV expert Richmond Alake to walk through the theory and practical components of computer vision and deep learning while also cultivating your Python knowledge. You’ll get explanations of widely used terminologies and learn to implement the techniques for and solutions to a typical CV image classification problem using Python, the TensorFlow machine learning library, and other standard data science packages, such as NumPy and pandas.
Hands-on learning with Jupyter notebooks
All exercises and labs are provided as Jupyter notebooks—interactive documents that combine live code, equations, visualizations, and narrative text. There's nothing to install or configure; just click a link and get started! And you can revisit them anytime after class ends to practice and refine your skills.
What you’ll learn and how you can apply it
By the end of this live online course, you’ll understand:
- Computer vision and its application in practical environments
- The fundamental components of deep learning as a machine learning field
- How to leverage deep learning to solve computer vision tasks
- The structural components of deep neural networks and convolutional neural networks
And you’ll be able to:
- Implement solutions to common computer vision tasks
- Use machine learning libraries to implement deep learning solutions
- Build a deep neural network that classifies images
- Build a convolutional neural network (AlexNet) that classifies images
- Visualize the neural network training process using TensorBoard
This live event is for you because...
- You’re a machine learning practitioner who wants to upskill.
- You want to explore computer vision, one of the prevalent fields within machine learning and AI.
- You work with Python and want to understand its use in machine learning and deep learning.
- You want to build a solid foundation for more advanced studies.
- You want to understand the technology behind modern AI tools and applications.
Prerequisites
- Experience developing small programs with Python
- A basic understanding of machine learning concepts
- Familiarity with machine learning tools and libraries such as TensorFlow, NumPy, pandas, and Matplotlib
- Familiarity with the Jupyter Notebook or JupyterLab
Recommended preparation:
- Watch Introduction to Python (video)
Recommended follow-up:
- Read Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow, second edition (book)
Schedule
The time frames are only estimates and may vary according to how the class is progressing.
Introduction to computer vision and deep learning (60 minutes)
- Presentation: What are neural networks?; deep neural networks; convolutional neural networks; CNN architectures
- Group discussion: Have you ever implemented a neural network from scratch?; What computer vision application examples can you identify?
- Q&A
- Break
Solving computer vision problems with deep learning models: Part I (60 minutes)
- Presentation: Keras sequential class and activation functions; viewing details of models; loading the Fashion-MNIST dataset; partitioning dataset into test/training/validation split; dataset visualization with Matplotlib; training a deep neural network (learning rate, learning rate schedules, loss functions, optimizers); viewing model training insights with TensorBoard; evaluating a trained model
- Jupyter notebooks: Implement a Deep Neural Network; Classify an Image with a DNN
- Q&A
- Break
Solving computer vision problems with deep learning models: Part II (50 minutes)
- Presentation: Introduction to AlexNet; layers (convolutional, batch normalization, max pooling, dropout); model implementation with Keras Sequential class API; loading the CIFAR-10 dataset; partitioning dataset into test/training/validation split; preprocessing datasets; training AlexNet (viewing model details and model training insights with TensorBoard); evaluating a trained model
- Jupyter notebooks: Implement AlexNet (CNN) from Scratch; Classify an Image with AlexNet
- Wrap-up and Q&A (10 minutes)
Your Instructor
Richmond Alake
Richmond Alake is a Machine Learning and Computer Vision Engineer currently working at LoveShark, a mobile gaming studio. He's been involved in the technology field for over five years and has worked for large conglomerates, financial institutions and small startups.
Richmond believes in using technology to solve everyday problems and has built several websites and mobile applications. He's written over 100 articles with over a million views on Machine Learning topics. He has produced several popular articles like How To Read Research Papers, Took A Masters In Machine Learning and What I learnt From Taking A Masters In Computer Vision.
Richmond believes in the robust application of machine learning to everyday problems and is currently heading several projects that leverage machine learning algorithms and deep learning models to solve ergonomic and social network related problems.