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.
Solution: Experiment with hyperparameters
From the course: TensorFlow 2.0: Working with Images
Solution: Experiment with hyperparameters
- So let's head over to our first challenge. So we want to be able to include input nodes of 1024 and 512 nodes. So I'm going to reuse these two lines where we have 128 nodes and 64 nodes and let's replace these with 1024 nodes here, and 512 nodes here. Now, just so happens that with this iteration of my model, I had an accuracy of around 45%. So let's see how that compares with the changes that we're going to make here. So let me go ahead and rerun the cell. So our model has now finished training. You can see that our model now has an accuracy of around 49%, which is better than it had before. So it looks like including these internal nodes of 1024 and 512 helps to improve the accuracy of the model. Now let's head over to the second part of the challenge. And what I'm going to do, is I'm going to take the first section and I'm going to include this in a new cell. So let's take a look at the dimensions of train dataset. If…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.