English | MP4 | AVC 1280×720 | AAC 44KHz 2ch | 0h 54m | 300 MB
This video teaches you how to build a powerful image classifier in just minutes using convolutional neural networks and PyTorch. It reviews the fundamental concepts of convolution and image analysis; shows you how to create a simple convolutional neural network (CNN) with PyTorch; and demonstrates how using transfer learning with a deep CNN to train on image datasets can generate state-of-the-art performance. The course is designed for the software engineer looking to get started with deep learning and for the AI researcher with TensorFlow or Theano experience who wants a smooth transition into PyTorch. Prerequisites include an understanding of algebra, basic calculus, and basic Python skills. Learners should download and install PyTorch before starting class.
- Learn how to build a powerful image classifier in minutes using PyTorch
- Explore the basics of convolution and how to apply them to image recognition tasks
- Learn how to do transfer learning in conjunction with powerful pretrained models
- Gain experience using powerful deep learning models for image recognition tasks
Table of Contents
01 Course Introduction
02 The Curse of Dimensionality with Traditional Feed Forward Networks
03 Exploiting Locality and Stationarity of Data with Convolutions
04 CNNs for Image Processing
05 Simple CNN for MNIST classification using PyTorch
06 Popular CNN Architectures for Image Recognition
07 Using Popular CNNs in PyTorch
08 CNNs for Document Classification using PyTorch
09 Conclusion
Resolve the captcha to access the links!