List of Listings

Chapter 2. TensorFlow essentials

Listing 2.1. Computing the inner product of two vectors without using a library

Listing 2.2. Computing the inner product using NumPy

Listing 2.3. Different ways to represent tensors

Listing 2.4. Creating tensors

Listing 2.5. Using the negation operator

Listing 2.6. Using a session

Listing 2.7. Using the interactive session mode

Listing 2.8. Logging a session

Listing 2.9. Using a variable

Listing 2.10. Saving variables

Listing 2.11. Loading variables

Listing 2.12. Defining the average update operator

Listing 2.13. Running iterations of the exponential average algorithm

Listing 2.14. Filling in missing code to complete the exponential average algorithm

Listing 2.15. Annotating with a summary op

Listing 2.16. Writing summaries to view in TensorBoard

Chapter 3. Linear regression and beyond

Listing 3.1. Visualizing raw input

Listing 3.2. Solving linear regression

Listing 3.3. cost functionUsing a polynomial model

Listing 3.4. Splitting the dataset into testing and training sets

Listing 3.5. Evaluating regularization parameters

Listing 3.6. Parsing raw CSV datasets

Chapter 4. A gentle introduction to classification

Listing 4.1. Using linear regression for classification

Listing 4.2. Executing the graph

Listing 4.3. Measuring accuracy

Listing 4.4. Linear regression failing miserably for classification

Listing 4.5. Using one-dimensional logistic regression

Listing 4.6. Setting up data for two-dimensional logistic regression

Listing 4.7. Using TensorFlow for multidimensional logistic regression

Listing 4.8. Visualizing multiclass data

Listing 4.9. Setting up training and test data for multiclass classification

Listing 4.10. Using softmax regression

Listing 4.11. Executing the graph

Chapter 5. Automatically clustering data

Listing 5.1. Traversing a directory for data

Listing 5.2. Representing audio in Python

Listing 5.3. Obtaining a dataset for k-means

Listing 5.4. Implementing k-means

Listing 5.5. Organizing data for segmentation

Listing 5.6. Segmenting an audio clip

Listing 5.7. Setting up the SOM algorithm

Listing 5.8. Defining how to update the values of neighbors

Listing 5.9. Getting the node location of the closest match

Listing 5.10. Generating a matrix of points

Listing 5.11. Running the SOM algorithm

Listing 5.12. Testing the implementation and visualizing the results

Chapter 6. Hidden Markov models

Listing 6.1. Defining the HMM class

Listing 6.2. Creating a helper function to access emission probability of an observation

Listing 6.3. Initializing the cacheee

Listing 6.4. Updating the cache

Listing 6.5. Defining the forward algorithm given an HMM

Listing 6.6. Defining the HMM and calling the forward algorithm

Listing 6.7. Adding the Viterbi cache as a member variable

Listing 6.8. Defining an op to update the forward cache

Listing 6.9. Defining an op to update the back pointers

Listing 6.10. Defining the Viterbi decoding algorithm

Listing 6.11. Running the Viterbi decode

Chapter 7. A peek into autoencoders

Listing 7.1. Python class schema

Listing 7.2. Using name scopes

Listing 7.3. Autoencoder class

Listing 7.4. Training the autoencoder

Listing 7.5. Testing the model on data

Listing 7.6. Using your Autoencoder class

Listing 7.7. Batch helper function

Listing 7.8. Batch learning

Listing 7.9. Loading images

Listing 7.10. Reading from the extracted CIFAR-10 dataset

Listing 7.11. Reading all CIFAR-10 files to memory

Listing 7.12. Converting CIFAR-10 image to grayscale

Listing 7.13. Setting up the autoencoder

Chapter 8. Reinforcement learning

Listing 8.1. Importing relevant libraries

Listing 8.2. Helper function to get prices

Listing 8.3. Helper function to plot the stock prices

Listing 8.4. Get data and visualize it

Listing 8.5. Defining a superclass for all decision policies

Listing 8.6. Implementing a random decision policy

Listing 8.7. Using a given policy to make decisions, and returning the performance

Listing 8.8. Running multiple simulations to calculate an average performance

Listing 8.9. Defining the decision policy

Listing 8.10. Implementing a more intelligent decision policy

Chapter 9. Convolutional neural networks

Listing 9.1. Loading images from a CIFAR-10 file in Python

Listing 9.2. Cleaning data

Listing 9.3. Preprocessing all CIFAR-10 files

Listing 9.4. Using the cifar_tools helper function

Listing 9.5. Visualizing images from the dataset

Listing 9.6. Generating and visualizing random filters

Listing 9.7. Using a session to initialize weights

Listing 9.8. Showing convolution results

Listing 9.9. Visualizing convolutions

Listing 9.10. Running the maxpool function to subsample convolved images

Listing 9.11. Setting up CNN weights

Listing 9.12. Creating a convolution layer

Listing 9.13. Creating a max-pool layer

Listing 9.14. The full CNN model

Listing 9.15. Defining ops to measure the cost and accuracy

Listing 9.16. Training the neural network by using the CIFAR-10 dataset

Chapter 10. Recurrent neural networks

Listing 10.1. Importing relevant libraries

Listing 10.2. Defining a class and its constructor

Listing 10.3. Defining the RNN model

Listing 10.4. Training the model on a dataset

Listing 10.5. Testing the learned model

Listing 10.6. Training and testing on dummy data

Listing 10.7. Loading data

Listing 10.8. Modifying the test function to pass in the session

Listing 10.9. Generate training data

Chapter 11. Sequence-to-sequence models for chatbots

Listing 11.1. Setting up constants and placeholders

Listing 11.2. Making a simple RNN cell

Listing 11.3. Stacking two RNN cells

Listing 11.4. Using MultiRNNCell to stack multiple cells

Listing 11.5. Defining a lookup table of scalars

Listing 11.6. Defining a lookup table of 4D vectors

Listing 11.7. Defining a lookup table of tensors

Listing 11.8. Looking up the embeddings

Listing 11.9. Extracting character vocab

Listing 11.10. Defining hyperparameters

Listing 11.11. Listing placeholders

Listing 11.12. Helper functions to build RNN cells

Listing 11.13. Encoder embedding and cell

Listing 11.14. Preparing input sequences to the decoder

Listing 11.15. Decoder embedding and cell

Listing 11.16. Decoder output (training)

Listing 11.17. Decoder output (inference)

Listing 11.18. Cost function

Listing 11.19. Calling an optimizer

Listing 11.20. Training the model

Chapter 12. Utility landscape

Listing 12.1. Importing relevant libraries

Listing 12.2. Generating dummy training data

Listing 12.3. Placeholders

Listing 12.4. Hidden layer

Listing 12.5. Output layer

Listing 12.6. Loss and optimizer

Listing 12.7. Preparing a session

Listing 12.8. Training the network

Listing 12.9. Preparing test data

Listing 12.10. Visualize results

Listing 12.11. Importing libraries

Listing 12.12. Preparing the training data

Listing 12.13. Placeholder

Listing 12.14. Preparing the session

Listing 12.15. Loading the VGG16 model

Listing 12.16. Preparing data for ranking

Listing 12.17. Training the ranking network

Listing 12.18. Preparing image sequences from video

Listing 12.19. Computing the utility of images

Listing 12.20. Visualizing utility scores