The importer for the TensorFlow models would enable you to import a pretrained TensorFlow models and weights. You can then use this model for prediction or transfer learning. Alternatively, you can import layer architecture as a Layer array or a LayerGraph object. You can then train this model.

Opening the kerasimporter.mlpkginstall file from your operating system or from within MATLAB will initiate the installation process for the release you have.

The capabilities for importing TensorFlow SavedModel format via using importTensorFlowNetwork and importTensorFlowLayers are added in R2021a. These functions are compatible with TensorFlow 2 only. If you're working with R2017b through R2020b, you can use importKerasNetwork and importKerasLayers functions to import models in the HDF5 format. It is highly recommended that you install the latest version of the support package and use it with the MATLAB R2021a release for the best results.

The following links will help you get started with examples:
Import pretrained TensorFlow network ://www.tianjin-qmedu.com/help/deeplearning/ref/importtensorflownetwork.html
Import layers from TensorFlow network://www.tianjin-qmedu.com/help/deeplearning/ref/importtensorflowlayers.html