# Introduction to TensorFlow Contents

### Introduction

Introduction to different types of AI models, and the focus of the class.

### Tensorflow Linear Regression

This section discusses Placeholders, Variables, Sessions, Running sessions, optimization, and calculating error

### Exercise 1

Experiment with a predefined TensorFlow model to change learning rate, epochs and more

### Tensorflow Classification

This section discusses Softmax definition, Cross Entropy definition, error propagation, Tensorflow Optimizers, Learning rate, and epochs

### Exercise 2

Experiment with a predefined TensorFlow MNIST classifier

### Tensorflow Deep Networks

This section discusses Tensorflow activation functions, Tensorflow differentiation, Tensorflow error propagation, Tensorflow Deep Network optimization, and Deep Network descriptions

### Exercise 3

Experiment with a predefined TensorFlow Deep MNIST classifier including adding extra layers

### Visualizing Model Operation

This section discusses Tensorboard, adding summaries, adding histograms, adding graphs, interpreting results

### Exercise 4

Use Tensorboard to visualize the internal workings of a TensorFlow model during training and evaluation

### Convolutional Neural Networks

This section discusses Convolutional filters, feature maps, convolutional layers, pooling layers, fully connected layers, stride, padding, constructing CNN networks, training CNN networks

### Exercise 5

Experiment with a predefined TensorFlow CNN model. Change hyper parameters and see what happens

### TensorFlow Estimators

This section discusses TensorFlow Built-In Estimators, Estimator behavior, Why use estimators?

### Exercise 6

Experiment with a predefined TensorFlow Estimator model

### Transfer Learning

What makes a good data set, balanced data sets, distinct data sets, non-conflicting data, ImageNet, Inception-V3, transfer learning description, transfer learning operation, transfer learning

### Exercise 7

Experiment with using TensorFlow scripts to retrain a TensorFlow Inception-V3 network using different training data