Distributed Computing & AI with Spark

Overview

You’ll get hands-on experience in building, training, and customizing models using TensorFlow and Keras. Starting from implementing simple networks by hand, this course builds to making use of advanced pre-trained models with transfer learning.

You’ll complete a practical mini-project on image classification, with the help of guidance during office hours.

Syllabus

• Introduction to Distributed Computing and AI
Overview of distributed computing principles
Introduction to Spark for big data processing
• Introduction to TensorFlow and Keras
Setting up the TensorFlow environment
Basics of neural networks with TensorFlow
• Building Simple Neural Networks
Implementing neural networks by hand using Keras
Understanding layers, activation functions, and loss functions
• Advanced TensorFlow Techniques
Customizing models with advanced TensorFlow techniques
Implementing callbacks, optimizers, and metrics
• Introduction to Spark for AI
Integrating Spark with TensorFlow
Using Spark for distributed model training
• Transfer Learning
Understanding pre-trained models
Fine-tuning models with Keras
• Practical Mini-Project: Image Classification
Problem definition and data exploration
Building and training an image classifier with TensorFlow
Evaluating model performance and optimizing
• Hands-On Labs and Exercises
Regular hands-on sessions to reinforce learning
Step-by-step guidance for implementing concepts taught
• Office Hours
Scheduled sessions for project guidance and Q&A
• Final Project and Presentation
Complete the image classification mini-project
Present project findings and lessons learned
• Course Conclusion and Next Steps
Recap of key concepts
Recommendations for further learning and exploration in AI and distributed computing

Subjects

Data Science