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.

Using ModelCheckpoint

Using ModelCheckpoint

From the course: TensorFlow 2.0: Working with Images

Start my 1-month free trial Buy for my team

Using ModelCheckpoint

“

- [Instructor] ModelCheckpoints allow you to save your model regularly throughout the training process. So this is usually at the end of each training epoch. The model weights and biases are finalized and saved as a weight file. Use ModelCheckpoint to save the model only if it is improved from the previous epochs. So the last saved model is going to be the best. Let's head over to our Colab notebook to see how ModelCheckpoints work. So I'm going to head over to runtime. I want to make sure that I'm still at GPU as my hardware accelerator. And then I'm going to go ahead and run all the cells in this notebook. So what we're going to do is to download our CIFAR-10 dataset. We have the images here. Each batch will contain 128 images and each training batch is fed into the model during the training process to enable incremental updates to weights and biases. So to help with the understanding of how we might go about…

Contents