From the course: Deep Learning with Python: Convolutional Neural Networks
Join today to access over 24,700 courses taught by industry experts.
Inception networks - Python Tutorial
From the course: Deep Learning with Python: Convolutional Neural Networks
Inception networks
- [Instructor] Inception Networks, also referred to as GoogLeNet, were introduced by researchers at Google in 2014 with the aim of building a deep convolutional neural network that achieved high accuracy while maintaining low computational costs. Unlike traditional CNNs that stack layers sequentially, Inception introduced a novel multi-path architecture. Each Inception block processes the input through multiple parallel paths using a one-by-one, three-by-three, and five-by-five convolutions, as well as max pooling. The results from each path are then concatenated along the channel dimension. This structure allows the network to analyze features at multiple spatial scales simultaneously, making it more efficient and effective at learning image representations. One of the key advantages of Inception Networks is their ability to achieve high accuracy with fewer parameters. For instance, GoogLeNets use approximately 4 million parameters to reach a top five error rate of 6.7% on ImageNet…
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.