Index

Search in book...
Toggle Font Controls
Create new playlist

Name your new playlist

Playlist description (optional)
Sign In

Email address

Password

Forgot Password?

or

Continue with Facebook

Continue with Google
Sign Up

Full Name

Email address

Confirm Email Address

Password

or

Continue with Facebook

Continue with Google

Previous Chapter

Appendix. Installation

List of Figures

Index

[SYMBOL][A][B][C][D][E][F][G][H][I][J][K][L][M][N][O][P][Q][R][S][T][U][V][W][Y]

SYMBOL

+ (plus symbol) 1D (one-dimensional) tensor 500-by-500 tensor

A

accuracy ACTION_DROP, responding to actions activation function AdamOptimizer ADC (analog-to-digital converter) add_summary method amounts vector arbitrary functions artificial neural network AUC (area-under-curve) value audio, extracting features from augment data autoencoders, 2^nd, 3^rd applying batch training neural networks working with images automatic differentiation capabilities autonomous agent

B

BasicLSTMCell batch learning batch training Bayesian networks best-fit curve, 2^nd, 3^rd bias binary classifier, 2^nd black box BMU (best matching unit) Boolean variables, Python BregmanToolkit

C

cells, evaluating centroid chatbots, seq2seq models architecture of classifications gathering dialogue data RNNs vector representation of symbols chromagram CIFAR-10 web page cifar_tools.py file classification formal notation for measuring performance of classification algorithm accuracy precision and recall ROC curve multivariate classification one-versus-all approach for one-versus-one approach for softmax regression for overview real-world application of using linear regression for using logistic regression for one-dimensional logistic regression two-dimensional logistic regression classifiers, training clustering, 2^nd extracting features from audio k-means clustering loading files in TensorFlow real-world application of segmentation SOM CNNs (convolutional neural networks), 2^nd code performance, trouble-shooting visualizing as graph writing in Jupyter configurations, session confusion matrix, 2^nd constant value contextual information continuous outputs conv2d function, 2^nd convex convolution layer convolutional neural networks applying implementing in TensorFlow measuring performance training classifiers improving performance of neural networks, disadvantages of overview preparing images convolving using filters generating filters max-pooling convolving images, using filters Cornell Movie Dialogue corpus correct_prediction correspondence problem cost function, 2^nd, 3^rd, 4^th, 5^th, 6^th cross-entropy loss CSV datasets curr_value curse of dimensionality

D

data cleaning clustering extracting features from audio k-means clustering loading files in TensorFlow real-world application of segmentation SOM dialogue, gathering encoding higher-dimensional linear fit loading overfitting, 2^nd underfitting visualizing using TensorBoard implementing moving averages visualizing moving averages. See time-series data. data representation, 2^nd data split dataflow graph, 2^nd Data.gov data_loader.py file datasets, CSV deciding boundary decoder_input_embedded decoder_input_seq, 2^nd decoding deeper architecture DeepRL denoising autoencoder dependent variables dimensionality reduction discount factor discrete Fourier transform discrete outputs distance metrics DistBelief Docker tool installing on Linux installing on macOS installing on Windows to install TensorFlow using docker-machine ip command dot product, 2^nd dtype (data type) dummy data, 2^nd dummy variables

E

early stopping edges, 2^nd E-M algorithm embedding images embedding_lookup function embed_sequence function emission matrix, 2^nd emission probability matrix encoder_state variable encoding <EOS> (end-of-sequence) symbol epoch epsilon parameter Euclidian distance eval() function executing a forward step exploration vs. exploitation, 2^nd

F

face detection and pose estimation, databases for fast Fourier transform FDDB (Face Detection Data Set and Benchmark) feature engineering, 2^nd feature vector, 2^nd, 3^rd features distance metrics extracting from audio overview feed_dict argument files, loading in TensorFlow filters convolving images using generating FN (false negative) forward algorithm Fourier transform FP (false positive) freq variable frequency domain Frossard, Davi future time-steps

G

game playing GAN (generative adversarial networks) Gaussian distribution, 2^nd generating filters genetic algorithms global_variables_initializer op <GO> symbol go_prefixes matrix gradient clipping gradient descent graphs, visualizing code as greedy policy

H

Hardware-Assisted Virtualization Detection Tool, Microsoft helper functions hidden layers, 2^nd hidden neurons, 2^nd HMMs (hidden Markov models) forward algorithm Markov model modeling DNA modeling images modeling video not-so-interpretable models overview of real-world application of Viterbi decoding algorithm hyperbolic tangent function hyperparameters, 2^nd

I

identity ImageNet dataset images embedding loading preparing convolving using filters generating filters max-pooling ranking working with img placeholder imitation learning independent variables inductive learning inference inference_logits op initial probabilities initial probability vector input_dim installing Matplotlib TensorFlow by using Docker installing Docker on Linux installing Docker on macOS installing Docker on Windows using Docker int32 type interactive session mode interpretable models

J

jerk Jupyter Notebook, writing code in

K

Kaggle k-means algorithm k-means clustering

L

L0 norm L1 norm L2 norm Large Movie Review Dataset libraries, testing LibriSpeech linear decision boundary linear functions linear model linear regression for classification formal notation for polynomial model real-world application of regularization simple regression model solving L-infinity norm Linux operating system, installing Docker on L-N norm loading variables load_sentences function load_series function logistic regression, for classification one-dimensional two-dimensional LSTM (Long Short-Term Memory) LSTMCell class

M

machine learning fundamentals of inference learning parameters types of reinforcement learning supervised learning unsupervised learning macOS (operating system), installing Docker on main function make_cell function Manhattan distance maps, SOM Markov model Matplotlib library, installing matrix, 2^nd max function, 2^nd max-pooling technique, 2^nd MDP (Markov decision process) metaparameter Moby model, 2^nd, 3^rd moving averages implementing visualizing multiclass classifier MultiRNNCell multivariate classification one-versus-all approach for one-versus-one approach for softmax regression for multivariate regression

N

name scopes negative operator, 2^nd negative value neural networks applying, 2^nd contextual information disadvantages of implementing implementing in TensorFlow measuring performance training classifiers improving performance of overview, 2^nd predictive model for time-series data preparing images convolving using filters generating filters max-pooling NLP (natural language processing), 2^nd nodes, 2^nd nominal values nonconvex function nonlinear functions normal distribution, 2^nd num_iters NumPy array, 2^nd

O

one-dimensional (1D) tensor one-hot encoding, 2^nd, 3^rd one-versus-all approach, for multivariate classification one-versus-one approach, for multivariate classification operations, defining operators creating executing with sessions setting session configurations visualizing code as graph optimal point optimal policy ordinal type, values outliers output layers, 2^nd overfitting, 2^nd

P

parameters, 2^nd parameter-space PCA (principle component analysis) pixel intensity placeholders, 2^nd, 3^rd plus symbol (+) policies, in RL polynomial regression model precision predictive models, for time-series data preference models probability matrix

Q

Q-function, 2^nd, 3^rd Q-learning decision policy

R

ramp (ReLU) function random policy ranking, images reader.read function recall receiver operating characteristic curve. See ROC curve. recentering recurrent neural networks. See RNNs. regression for classification one-dimensional two-dimensional for multivariate classification formal notation for multivariate polynomial model real-world application of regularization simple regression model solving regressor regularization regularizing weights relu function representing tensors restore function rewards RL (reinforcement learning) applying, 2^nd implementing overview policies utilities rnn.dynamic_rnn function RNNs (recurrent neural networks), 2^nd applying contextual information implementing overview predictive model for time-series data robotics ROC (receiver operating characteristic) curve row-major order

S

saving variables scikit-learn library SciPy segmentation segment_size select_action method self.fwd self.obs_idx self-organizing map. See SOM. sentences Sentiment Labelled Sentences Data Set seq2seq (sequence-to-sequence) models architecture of classifications gathering dialogue data RNNs vector representation of symbols sequence_loss method SeriesPredictor class, 2^nd Session context session.run() statement sessions executing operators with setting session configura-tions visualizing code as graph logging show_weights function sig (sigmoid) function sigmoids simple_regression.py file simulated annealing slice function softmax function, 2^nd softmax_cross_entropy_with _logits function softmax.py file SOM (self-organizing map) split_data function stacked autoencoder, 2^nd stride length, 2^nd SummaryWriter supervised learning symbols, vector representation of

T

tb_files TED-LIUM temporal dependencies Tensor class TensorBoard dashboard, visualizing data using implementing moving averages visualizing moving averages TensorFlow library implementing convolutional neural networks in measuring performance training classifiers installing by using Docker installing Docker on Linux installing Docker on macOS installing Docker on Windows using Docker loading files in testing tensors, representing test function, 2^nd testing sets testing TensorFlow tf.add(x, y) operator tf.concat operator tf.constant operator tf.convert_to_tensor( ... ) function tf.div(x, y) operator tf.exp(x) operator tf.floordiv(x, y) operator tf.InteractiveSession() function tf.mod(x, y) operator tf.multiply(x, y) operator tf.nn.dropout function tf.nn.dynamic_rnn function tf.nn.softmax_cross_entropy _with_logits library tf.pow(x, y) operator tf.Session() function tf.sqrt(x) operator tf.subtract(x, y) operator tf.train.match_filenames_once operator tf.train.string_input_producer operator tf.truediv(x, y) operator three-dimensional vector time-series data, predictive model for Tinker TN (true negative) TP (true positive) train method, 2^nd training classifiers training dataset, 2^nd TrainingHelper train_op, 2^nd, 3^rd transition matrix, 2^nd trellis diagram Twitter Sentiment Analysis Dataset two-input network

U

underfitting unpickle function unsupervised learning update_avg operator utilities, in RL utility functions embedding images preference models ranking images

V

validation value of state Variable class variables loading saving using variance variational autoencoder vector representations, of symbols vectors VGG Face Dataset viterbi cache Viterbi decoding algorithm VoxForge

W

Windows operating system, installing Docker on WolframAlpha

Y

yahoo_finance library y-axis YouTube Faces Dataset

..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.

3.133.107.25