More talks in the program:
Frameworks for deep learning provide critical blocks for designing, training, and validating deep networks. TensorFlow has emerged as the industry’s most popular framework due to its highly scalable and flexible system architecture. However, as for an easy programming interface or debugging, it is still a pain. Tools like TF Debugger and TensorBoard have been introduced to make the user’s life simpler for debugging and visualizing deep learning.
In this session, we will try to understand deep learning concepts applied in computer vision and image processing. These tools help users understand the state of a learning algorithm, especially when it comes to the myriad of applications with images. We will examine a simple deep learning-based image recognition program, break it, and then learn to fix it effectively with the help of tools like TF Debugger and TensorBoard. We will also try to understand sophisticated deep learning concepts such as tuning the deep network by visualizing the effects on TensorBoard. We will learn how these handy tools can really make a big difference in deciphering the complexities of the algorithm. Avoid deep learning situations where you "don’t know what to do next!"