From the course: Building and Deploying Deep Learning Applications with TensorFlow
Unlock the full course today
Join today to access over 22,700 courses taught by industry experts or purchase this course individually.
Define the model structure - TensorFlow Tutorial
From the course: Building and Deploying Deep Learning Applications with TensorFlow
Define the model structure
- [Instructor] Alright, let's build a neural network with TensorFlow. Our training data set has nine input features, so we'll need nine inputs in our neural network. We can model that with a placeholder called X that holds nine values. Then, let's have three layers in their neural network that will train to find the relationship between the inputs and the output. There are many different types of layers you can use in the neural network, but we're going to use the most straightforward type, a fully connected neural network layer. That means that every node in each layer is connected to every node in the following layer. The first layer will have 50 nodes, the second layer will have 100 nodes, and the third layer will have 50 nodes again. To me, these layer sizes seem like a good starting point, but it's just a guess. Once the neural network is coded we can test out different layer sizes to see what layer size gives us the best accuracy. And since we are trying to predict a single…
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.