Description
Building a Neural Network from Zero course. This course shows how to build and train a complete neural network from scratch without using pre-written libraries to gain a thorough understanding of its internal mechanisms. This course is designed for those who want to advance their knowledge of neural networks and go beyond simple library usage to gain a fundamental understanding of how each component works. In this hands-on course, a Torchlight-like framework is created by hand so that participants can build, train, and evaluate neural networks. The learning process begins with basic principles such as numerical differentiation and gradient descent algorithms, and gradually develops a complete learning loop. By the end of the course, participants will have a comprehensive understanding of the neural network learning process and solidify their knowledge by applying their designed network to a practical project classifying the Fashion-MNES dataset. This course is suitable for machine learning engineers and programmers looking to gain foundational knowledge and hands-on experience building and customizing neural networks from scratch.
What you will learn
- Implementing a neural network from scratch:
- Implementing neural networks from scratch, including Forward Propagation and Backward Propagation.
- Mastering Optimization:
- Mastery of Gradient Descent, SGD with Momentum, and other Optimization Techniques.
- Custom component manufacturing:
- Create custom layers, activation functions, and loss functions without using external libraries.
- Practical applications:
- Applying a custom neural network to solve the Fashion-MNIST classification challenge.
- Core concepts:
- Numerical differentiation and three methods for calculating gradients.
- Gradient descent in 2D and multidimensional spaces.
- Implementation of Cross-Entropy Loss and activation functions such as Sigmoid.
- Initializing neural network weights using He and Xavier methods.
- Building a fully functional Feedforward Neural Network (FFNN) from scratch.
This course is suitable for people who:
- Beginners who want to understand how neural networks work in network infrastructure.
- Machine learning enthusiasts looking to deepen their knowledge through practical implementation.
- Developers who want to build custom neural network models from scratch.
- Students and professionals looking to strengthen their understanding of the core concepts of Deep Learning.
Course details
- Publisher: Udemy
- Instructor: Nick Ovchinnikov
- Training level: Beginner to advanced
- Training duration: 4 hours
Course syllabus as of 2025/4
Prerequisites for the Building a Neural Network from Zero course
- Basic knowledge of Python programming
- Familiarity with linear algebra concepts like vectors and matrices
- An interest in understanding neural networks at a fundamental level
Course images
Sample course video
Installation Guide
After Extract, view with your favorite player.
Subtitles: English
Quality: 1080p
Download link
Downloadly
Rapidgator link
File(s) password: www.downloadly.ir
File size
4.04 GB
