Tutorials at ADCOM 2024

Title: Deep Learning with Keras 3

Date: Tuesday, December 17th, 2024, Time: 8:30 AM – 4:30 PM

Introduces coding Deep Learning Models from scratch with Keras 3 that supports Tensorflow, Pytorch and Jax as backends.

PERSPECTIVE
Keras provides implementations of the most common building blocks of neural network architectures like layers of neurons, objective and activation functions, optimizers. It is a user-friendly, modular, and extensible open-source library (API) interface abstraction layer to simplify the use of TensorFlow for building neural network architectures, enabling fast experimentation through Python for efficient training and using scalable deep neural networks. Keras 3 is a full rewrite of Keras to be used as a low-level cross-framework language with unified codebase to develop custom components such as layers, models, or metrics that can be used for native workflows in TensorFlow, PyTorch or JAX

REQUIREMENTS
Basic Awareness of AI, ML, basic Python, is assumed from the participants who should carry their own laptops as there will be hands on demos. Colab will be used for practicals. No prior installation of software will be essential, apart from a standard browser. And, yes of course, ability to connect to Internet. Looks good.