TensorFlow Hub is a repository that contains trained machine learning models. These models are ready to be fine-tuned and deployed anywhere. The trained models such as BERT and Faster R-CNN can be reused with a few lines of code. It is an open-repository, which means it can be used and modified by anyone.
The tfhub.dev repository contains many pre-trained models. Some of them include text embeddings, image classification models, TF.js/TFLite models and so on.
It can be installed using the below code:
!pip install --upgrade tensorflow_hub
It can be imported into the working environment as shown in the below code:
import tensorflow_hub as hub