From the course: TensorFlow 2.0: Working with Images

Unlock the full course today

Join today to access over 23,200 courses taught by industry experts.

TensorFlow Hub

TensorFlow Hub

From the course: TensorFlow 2.0: Working with Images

Start my 1-month free trial Buy for my team

TensorFlow Hub

“

- [Announcer] Historically, deep learning engineers worked with complex models and trained them on large amounts of data, meaning that there are often significant costs involved in compute and in training of these models. So how can we use their work? Some of which might be TensorFlow without having to train these models from scratch. TensorFlow Hub is designed to solve this problem. It's enables transfer learning by making a variety of ML models, freely available as libraries or web API calls. Anyone can write just a single line of code to load the model and all models can be invoked via a simple web call and then the entire model is downloaded to your source codes runtime. And you don't even need to build a model yourself. This definitely saves development and training time. It also allows users to try out different models and build their own applications more quickly. Another benefit of transfer learning is that…

Contents