(Click Category to List Courses)
36 - ITC - Information Technology - Miscellaneous
ITC 232 - TensorFlow: The Complete Guide
| Code | Start Date | Duration | Venue | |
|---|---|---|---|---|
| ITC 232 | 21 July 2025 | 5 Days | Istanbul | Registration Form Link |
| ITC 232 | 15 September 2025 | 5 Days | Istanbul | Registration Form Link |
| ITC 232 | 10 November 2025 | 5 Days | Istanbul | Registration Form Link |
| ITC 232 | 01 December 2025 | 5 Days | Istanbul | Registration Form Link |
Course Description
TensorFlow 2 focuses on simplicity and ease of use, with updates like eager execution, intuitive higher-level APIs, and exible model building on any platform. Our TensorFlow 2 course is designed with all modern best practices for working with TensorFlow. The course will take you from an absolute beginner with TensorFlow, to becoming part of the growing Machine Learning industry. This course will guide you through how to use Google's latest TensorFlow 2 framework to create artificial neural networks for deep learning! This course aims to give you an easy-to-understand guide to the complexities of Google's TensorFlow 2 framework in a way that is easy to understand.
We'll focus on understanding the latest updates to TensorFlow and leveraging the Keras API (TensorFlow 2.0's official API) to quickly and easily build models. In this course, we will build models to forecast the future, classify, and translate texts artificially and much more!
This course is designed to balance theory and practical implementation, with complete Jupiter notebook guides of code and easy-to-reference slides and notes. We also have plenty of exercises to test your new skills along the way!
Perquisites:
- Patience, determination, and believe
- Completion of our available course AI and Deep Learning
- If the condition in 1. is not applicable then the learner must have a strong fundamental understanding of AI and Deep Learning.
Course Objectives
- Learn to use TensorFlow 2.0 for Deep Learning
- Leverage the Keras API to quickly build models that run on Tensorflow 2 Perform Image Classification with Convolutional Neural Networks
- Use Deep Learning imaging Forecast Time Series data with Recurrent Neural Networks
- Use Generative Adversarial Networks (GANs) to generate images
- Use deep learning for style transfer Generate text with RNNs and Natural Language Processing
- Serve Tensorflow Models through an API Use GPUs for accelerated deep learning
Who Should Attend?
- Developers who are interested to employ AI solutions
- Data Scientists, Engineers, and Software developers,
- Beginners to advanced learners who want to learn about deep learning and AI in Tensorflow 2.0
Course Details/Schedule
Day 1
- Structure of TensorFLow
- Academic review
- Perceptual framework
- Conceptual framework
- Setting up TensorFLow Environment
- Local installation
- Online Setup
- Eager execution
- Setup and usage,
- Dynamic Control Flow,
- Eager Training
- Advanced Automatic Differentiation
- Tensors
- Shapes
- Indexing
- Manipulating Shapes
- DType shapes
- Variables
- create variables
- Life-cycle, naming, and watching
Day 2
- Introduction to gradients and automatic differentiation
- Setup
- Computing gradients
- Gradients of non-scalar targets
- Control flow
- No gradient registered
- Introduction to graphs and tf.function
- Taking advantage of graphs
- Converting Python functions to graphs
- Seeing the speed-up
- When is a Function tracing?
- Introduction to modules, layers, and models Defining models and layers in TensorFlow
- Saving weights and functions
- Keras models and layers, and saving Keras models Basic training loops
- Solving machine learning problems
- Define the model
- The same solution, but with Keras
- Advanced automatic differentiation
- Controlling gradient recording
- Stop gradient flow with precision
- Jacobians
Day 3
- Keras
- The Sequential model
- Creating a Sequential model
- Feature extraction with a Sequential model Transfer learning with a Sequential model
- Training and evaluation with the built-in methods
- API overview: a first end-to-end example
- The compile method: specifying a loss, metrics, and an optimizer
- Making new Layers and Models via subclassing
- Setup
- The Layer class: the combination of state (weights) and some computation Save and load Keras models
- How to save and load a model
- Whole-model saving & loading
- Saving & loading only the model’s weights values
- Recurrent Neural Networks (RNN) with Keras
- Intro and setup
- Built-in RNN layers: a simple example
- RNN layers and RNN cells
- Performance optimization and CuDNN kernels
Day 4
- Data Input Pipelines
- tf.data: Build TensorFlow input pipelines
- Reading
- Patching
- Processing
- Iterating checkpoints
- Better performance with the tf.data API
- Setup
- Better Performance
- Optimization
- Analyze tf.data performance with the TF Profiler
- analysed workflow
- additional resources
- Checkpoints
- save model
Day 5
- Accelerators
- Setup
- Review and Evaluation
- Distributed training with TensorFlow
- Types of strategies
- Using tf.distribute.Strategy with tf.keras.Model.fit
- Using tf.distribute.Strategy with custom training loops
- GPU
- SETUP
- Logging device placement
- Use TPUs
- TPU initialization
- Manual device placement
- Distribution strategies
- Performance
- Better performance with tf.function
- Optimize TensorFlow performance using the Profiler
- Optimize TensorFlow GPU performance with the TensorFlow Profiler TensorFlow graph optimization with Grappler
