Index
A
- action-value function / Components of Reinforcement Learning
- action space / Components of Reinforcement Learning
- Adam optimization
- about / Understanding Adam Optimization
- reference link / Training and testing the model
- AI applications
- building / Recommendations for building AI applications
- in industries / AI applications in industries
- Amazon Transcribe / Speech-to-text frameworks and toolkits
- Apache Spark
- about / Introducing Apache Spark
- Resilient Distributed Dataset (RDD) / Introducing Apache Spark
- transformations and actions / Introducing Apache Spark
- DataFrames / Introducing Apache Spark
- Artificial Intelligence (AI)
- ethical consideration / Ethical considerations in AI
- auto-encoders / Understanding auto-encoders
B
- Bayes' rule / Understanding Bayes' rule
- Bayesian deep learning
- about / Understanding Bayesian deep learning
- rule, in neural networks / Bayes' rule in neural networks
- Bayesian inference / Introducing Bayesian inference
- Bayesian neural network
- building / Building a Bayesian neural network
- model, training / Defining, training, and testing the model
- model, defining / Defining, training, and testing the model
- model, testing / Defining, training, and testing the model
- Bayesian non-parametric models / Introducing Bayesian inference
- Bellman Optimality Equation / DQN for deep reinforcement learning
- binary class / Classification
- book scripts
- generating / Generating book scripts
- bottleneck features / Transfer learning
C
- capsule
- about / Understanding capsules
- working / How do capsules work?
- capsule networks (CapsNet)
- importance / Understanding the importance of capsule networks
- implementation / CapsNet implementation
- encoder / Understanding the encoder
- loss function, defining / Defining the loss function
- limitations / Limitations of capsule networks
- cartpole game
- simple policies, applying / Applying DQN to a game
- categorical cross entropy loss / Understanding categorical cross entropy loss
- classification / Machine learning, classification, and logistic regression
- Classification and Regression Trees (CART) / What is a decision tree?
- classification model evaluation metrics
- about / Classification Model Evaluation Metrics
- accuracy / Classification Model Evaluation Metrics
- precision / Classification Model Evaluation Metrics
- recall / Classification Model Evaluation Metrics
- clustering / Machine learning
- collaborative filtering / Collaborative filtering
- computation graph
- about / Computation graph
- lazy loading / The order of execution and lazy loading
- order of execution / The order of execution and lazy loading
- graph, executing across compute devices / Executing graphs across compute devices – CPU and GPGPU
- multiple graphs / Multiple graphs
- confusion matrix / Classification Model Evaluation Metrics
- content-based filtering
- about / Content-based filtering
- algorithms, advantages / Advantages of content-based filtering algorithms
- algorithms, disadvantages / Disadvantages of content-based filtering algorithms
- convolutional neural networks (CNNs) / Understanding the importance of capsule networks, Transfer learning
- credit card dataset
- reference link / Building a fraud detection model
- cross-domain relationships
- generator / Fundamental units of a DiscoGAN
- transposed convolution / Fundamental units of a DiscoGAN
- batch normalization / Fundamental units of a DiscoGAN
- Leaky ReLU / Fundamental units of a DiscoGAN
- discriminator / Fundamental units of a DiscoGAN
D
- data
- pre-processing / Pre-processing the data, Pre-processing the data
- data, pre-processing
- steps / Pre-processing the data
- datasets, testing
- URL, for downloading / Defining, training, and testing the model
- data types
- reference link / Tensors
- decision tree
- about / What is a decision tree?
- ensemble method, need for / Why do we need ensembles?
- ensemble method / Decision tree-based ensemble methods
- random forests / Random forests
- gradient boosting / Gradient boosting
- ensemble method, in TensorFlow / Decision tree-based ensembles in TensorFlow
- TensorForest Estimator / TensorForest Estimator
- TensorFlow boosted trees estimator / TensorFlow boosted trees estimator
- deep learning
- limitations / Limitations of deep learning
- Deep Q Networks (DQN)
- for deep reinforcement learning / DQN for deep reinforcement learning, Applying DQN to a game
- about / DQN for deep reinforcement learning
- Q-learning / DQN for deep reinforcement learning
- applying, to game / Applying DQN to a game
- deterministic / Components of Reinforcement Learning
- digits
- classifying, TensorFlow Lite used / Classifying digits using TensorFlow Lite
- DiscoGANs
- about / Understanding DiscoGANs
- fundamental units / Fundamental units of a DiscoGAN
- modeling / DiscoGAN modeling
- model, building / Building a DiscoGAN model
- discriminator, parameters
- layers / Fundamental units of a DiscoGAN
- activation / Fundamental units of a DiscoGAN
- normalizer / Fundamental units of a DiscoGAN
- stride / Fundamental units of a DiscoGAN
- distributed TensorFlow
- about / Understanding distributed TensorFlow
- deep learning / Deep learning through distributed TensorFlow
- synchronous approach / Deep learning through distributed TensorFlow
- asynchronous approach / Deep learning through distributed TensorFlow
- Docker containers
- URL / TensorFlow Serving
- dynamic routing algorithm / The dynamic routing algorithm
E
- Eager execution / Understanding TensorFlow.js
- encoder, CapsNet
- convolutional layer / Understanding the encoder
- primary caps layer / Understanding the encoder
- DigitCaps layer / Understanding the encoder
- episode / Components of Reinforcement Learning
- Evidence Lower Bound (ELBO)
- experience replay / DQN for deep reinforcement learning
F
- Fashion MNIST
- references / CapsNet for classifying Fashion MNIST images
- images, classifying with CapsNet / CapsNet for classifying Fashion MNIST images
- about / Training and testing the model
- fractional stride convolution/deconvolution / Fundamental units of a DiscoGAN
- fraud detection model
- building / Building a fraud detection model
- defining / Defining and training a fraud detection model
- training / Defining and training a fraud detection model
- testing / Testing a fraud detection model
- freeze graph / Converting TensorFlow model to TensorFlow Lite
- functions, TensorFlow probability (tpf)
- Tfp.distributions.categorical / Understanding TensorFlow probability, variational inference, and Monte Carlo methods
- probabilistic layers / Understanding TensorFlow probability, variational inference, and Monte Carlo methods
- Kullback-leibler (KL) divergence / Understanding TensorFlow probability, variational inference, and Monte Carlo methods
- variational inference / Understanding TensorFlow probability, variational inference, and Monte Carlo methods
- Monte Carlo method / Understanding TensorFlow probability, variational inference, and Monte Carlo methods
G
- Gaussian processes
- about / Introducing Gaussian processes
- kernels, selecting / Choosing kernels in GPs
- hyperparameters of kernel, selecting / Choosing the hyper parameters of a kernel
- applying, to stock market prediction / Applying GPs to stock market prediction
- Generative Adversarial Networks (GANs)
- training / Training GANs
- applications / Applications
- challenges / Challenges
- Deep convolutional GANs (DCGANs) / Challenges
- InfoGANs / Challenges
- Conditional GANs (cGANs) / Challenges
- generative model
- about / Understanding generative models
- variational autoencoders / Understanding generative models
- PixelRNN/PixelCNN / Understanding generative models
- generative adversarial networks / Understanding generative models
- Google Speech Commands Dataset
- about / Google Speech Commands Dataset
- reference link / Google Speech Commands Dataset
- GPflow
- reference link / Applying GPs to stock market prediction
- graph
- executing, across compute devices / Executing graphs across compute devices – CPU and GPGPU
- nodes, placing on compute devices / Placing graph nodes on specific compute devices
- simple placement / Simple placement
- dynamic placement / Dynamic placement
- soft placement / Soft placement
- GPU memory handling / GPU memory handling
H
- hybrid systems / Hybrid systems
K
- Kaldi
- Kepler dataset / Detecting exoplanets in outer space
- Keras
- used, for logistic regression / Logistic regression with Keras
- about / Logistic regression with Keras
- kernel
- about / Soft placement
- selecting, in Gaussian processes / Choosing kernels in GPs
- Kubernetes
- URL / TensorFlow Serving
L
- Lenet model
- reference link / Building a Bayesian neural network
- logistic regression
- about / Machine learning, classification, and logistic regression
- for binary classification / Logistic regression for binary classification
- for multiclass classification / Logistic regression for multiclass classification
- with TensorFlow / Logistic regression with TensorFlow
- with Keras / Logistic regression with Keras
- loss function, CapsNet
- margin loss / Defining the loss function
- reconstruction loss / Defining the loss function
M
- machine learning (ML)
- about / Machine learning, classification, and logistic regression, Machine learning
- supervised learning / Machine learning
- unsupervised learning / Machine learning
- classification / Classification
- logistic regression, for binary classification / Logistic regression for binary classification
- logistic regression, for multiclass classification / Logistic regression for multiclass classification
- margin loss / Defining the loss function
- matrix factorization
- about / Matrix factorization
- model, for Retailrocket recommendations / The matrix factorization model for Retailrocket recommendations
- Maximum a posteriori (MAP) / Understanding Bayesian deep learning
- Maximum Likelihood estimation (MLE)
- reference link / Understanding Bayesian deep learning
- MNIST dataset
- about / Logistic regression with TensorFlow
- reference link / Handwritten digits using TFoS
- model
- training / Training the model, Training and testing the model, Training the model
- testing / Training and testing the model
- model class
- designing / Defining the model
- model class, components
- input / Defining the model
- network definition / Defining the model
- sequence loss / Defining the model
- optimizer / Defining the model
- Module
- about / Understanding TensorFlow Hub
- composable / Understanding TensorFlow Hub
- reusable / Understanding TensorFlow Hub
- retrainable / Understanding TensorFlow Hub
- Monte Carlo methods / Understanding TensorFlow probability, variational inference, and Monte Carlo methods
- MsPacman-v0 / Creating a Pacman game in OpenAI Gym
- multiclass / Classification
- multiple graphs / Multiple graphs
N
- Natural Language Processing (NLP) / Understanding word embeddings
- network definition
- about / Defining the model
- LSTM cell, initializing / Defining the model
- word embedding / Defining the model
- LSTMs, building / Defining the model
- probability generation / Defining the model
- Neural Information Processing Systems (NIPS)
- reference link / TensorForest Estimator
- neural network architecture
- about / Neural network architecture
- feature extraction module / Feature extraction module
- deep neural network module / Deep neural network module
- neural network model
- for Retailrocket recommendations / The neural network model for Retailrocket recommendations
O
- object detection
- TensorFlowOnSpark (TFoS), used / Object detection using TensorFlowOnSpark and Sparkdl
- Sparkdl, used / Object detection using TensorFlowOnSpark and Sparkdl
- transfer learning / Transfer learning
- model, building / Building an object detection model
- OpenAI Gym
- about / OpenAI Gym
- reference link / OpenAI Gym
- Pacman game, creating / Creating a Pacman game in OpenAI Gym
- optimize for inference / Converting TensorFlow model to TensorFlow Lite
P
- Pacman game
- creating, in OpenAI Gym / Creating a Pacman game in OpenAI Gym
- posterior / Understanding Bayes' rule
- prior / Understanding Bayes' rule
R
- rank / Tensors
- recommendation systems / Recommendation systems
- reconstruction error / Understanding auto-encoders
- Rectified linear (ReLU) / Building a Bayesian neural network
- rectified linear units (ReLU) / Fundamental units of a DiscoGAN
- recurrent / Understanding recurrent neural networks
- recurrent neural networks (RNNs)
- about / Understanding recurrent neural networks
- Long Short-Term Memory (LSTMs) / Understanding recurrent neural networks
- Gated recurrent units (GRU) / Understanding recurrent neural networks
- Peephole LSTMs / Understanding recurrent neural networks
- reinforcement learning
- about / Machine learning, Reinforcement learning, Reinforcement learning versus supervised and unsupervised learning
- versus supervised learning / Reinforcement learning versus supervised and unsupervised learning
- versus unsupervised learning / Reinforcement learning versus supervised and unsupervised learning
- components / Components of Reinforcement Learning
- remote procedure call (RPC) / TensorFlow Serving
- Retailrocket dataset
- about / Introducing the Retailrocket dataset
- reference link / Introducing the Retailrocket dataset
- exploring / Exploring the Retailrocket dataset
- Retailrocket recommendations
- used, for matrix factorization model / The matrix factorization model for Retailrocket recommendations
- used, for neural network model / The neural network model for Retailrocket recommendations
S
- sample images
- reconstructing / Reconstructing sample images
- scalable / Introducing Apache Spark
- sentiment analysis
- model, building / Building the sentiment analysis model, Building the model
- model, pre-processing data / Pre-processing data
- model, data preprocessing / Pre-processing data
- shape / Tensors
- Singular Value Decomposition (SVD) / Matrix factorization
- Sparkdl
- used, for object detection / Object detection using TensorFlowOnSpark and Sparkdl
- interface / Understanding the Sparkdl interface
- squared exponential (SE) kernel / Choosing kernels in GPs
- squashing / How do capsules work?
- state-value function / Components of Reinforcement Learning
- state space / Components of Reinforcement Learning
- stochastic / Components of Reinforcement Learning
- stock market prediction
- Gaussian processes, applying to / Applying GPs to stock market prediction
- stock price prediction model
- creating / Creating a stock price prediction model
- obtained results / Understanding the results obtained
- supervised learning
- about / Machine learning, Reinforcement learning versus supervised and unsupervised learning
- versus reinforcement learning / Reinforcement learning versus supervised and unsupervised learning
- Support Vector Machines (SVM) / Defining the loss function
T
- TensorFlow
- about / What is TensorFlow?
- low-level API / What is TensorFlow?
- high-level API / What is TensorFlow?
- decision tree, ensemble method / Decision tree-based ensembles in TensorFlow
- Gradient Boosted Trees model, building for exoplanets detection / Building a TFBT model for exoplanet detection
- implementing, in production / Implementing TensorFlow in production
- TensorFlow.js
- about / Understanding TensorFlow.js
- model, executing on browser / Running the model on a browser using TensorFlow.js
- TensorFlow Boosted Trees (TFBT) / TensorFlow boosted trees estimator
- TensorFlow boosted trees estimator / TensorFlow boosted trees estimator
- TensorFlow core
- about / The TensorFlow core
- tensors / Tensors
- constants / Constants
- operations / Operations
- placeholders / Placeholders
- creating, from Python objects / Tensors from Python objects
- variables / Variables
- generating, from library functions / Tensors generated from library functions
- variables, with tf.get_variable() / Obtaining variables with the tf.get_variable()
- TensorFlow Extended (TFX)
- about / TensorFlow Extended
- analyze data / TensorFlow Extended
- transform / TensorFlow Extended
- train TF estimator / TensorFlow Extended
- analyze model / TensorFlow Extended
- serve model / TensorFlow Extended
- TensorFlow Hub / Understanding TensorFlow Hub
- TensorFlow Lite
- about / What is TensorFlow Lite?
- advantages / What is TensorFlow Lite?
- used, for classifying digits / Classifying digits using TensorFlow Lite
- data, pre-processing / Pre-processing data and defining the model
- model, defining / Pre-processing data and defining the model
- TensorFlow model, converting / Converting TensorFlow model to TensorFlow Lite
- TensorFlowOnSpark (TFoS)
- about / Learning about TensorFlowOnSpark
- architecture / Understanding the architecture of TensorFlowOnSpark
- Spark RDD / Understanding the architecture of TensorFlowOnSpark
- TensorFlow QueueRunners / Understanding the architecture of TensorFlowOnSpark
- usage / Deep delving inside the TFoS API
- used, for handwritten digits / Handwritten digits using TFoS
- used, for object detection / Object detection using TensorFlowOnSpark and Sparkdl
- TensorFlow Probability (tfp)
- TensorFlow serving
- about / TensorFlow Serving
- C++ libraries / TensorFlow Serving
- binaries / TensorFlow Serving
- hosted services / TensorFlow Serving
- advantages / TensorFlow Serving
- TensorForest Estimator / TensorForest Estimator
- text-generating model
- Text Frameworks
- tf.cond
- reference link / Training and testing the model
- Toolkits
- trajectory / Components of Reinforcement Learning
- transfer learning / Object detection using TensorFlowOnSpark and Sparkdl, Transfer learning
U
- unsupervised learning
- about / Machine learning, Reinforcement learning versus supervised and unsupervised learning
- versus reinforcement learning / Reinforcement learning versus supervised and unsupervised learning
V
- variational inference
W
- WebGL / Understanding TensorFlow.js
- white noise kernel / Choosing kernels in GPs
- word embedding
- about / Understanding word embeddings
- reference link / Understanding word embeddings