O'Reilly logo

Hands-On Serverless Deep Learning with TensorFlow and AWS Lambda by Rustem Feyzkhanov

Stay ahead with the world's most comprehensive technology and business learning platform.

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, tutorials, and more.

Start Free Trial

No credit card required

Pre-trained networks

Although training neural networks may require large processing power and a lot of data, deploying them can be done using a simple CPU. In this way, we can say that deploying a deep learning model is close to using an external library in your code. Secondly, there's a large community of people and companies that open source their pre-trained neural networks, which means that you can freely use them.

There are two instances where using pre-trained neural networks can be extremely convenient:

  • The first case is when your task has already been solved. For example, if you want to conduct image captioning with X classification, then you can use already existing neural networks.
  • The second case is when your task is fairly different ...

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, interactive tutorials, and more.

Start Free Trial

No credit card required