Index

A

Activation functions
bilinear function
definition
hyperbolic tangent function
leaky ReLU
linear function
log sigmoid transfer function
PyTorch vs . TensorFlow
ReLU
sigmoid function
visualization, shape of
Adadelta
Adagrad
Adam optimizer
Algorithmic classification
Alternate data collection
Artificial neural network (ANN)
Autoencoders
architecture
clustering model
encoded data, 3D plot
features
hyperbolic tangent function
layer
MNIST dataset
torchvision library
Autograd

B

Bernoulli distribution
Beta distribution
Bilinear function
Binomial distribution

C

Central processing units (CPUs)
Computational graph
Computational linguistics
Continuous bag of words (CBOW)
example
representation
vectors embedding
Continuous uniform distribution
Convolutional neural network (CNN)
architecture
computational process
hyperparameters
loader functionality
MNIST dataset
net1 object
pickle file format
pooling layer
prediction services
predictions process
restore function
restore_net() function
save function
test accuracy
training loss

D, E

Data mining
Deep learning models
batch size
batch training
CPU environment
loss function
online training
hyperparameters
learning rate
parallel data training
sequential neural network
Deep neural network (DNN)
Discrete probability distribution
Double exponential distribution

F

Facebook’s artificial intelligence

G

GloVe
Gradient computation
Gradient descent algorithm
Graphics processing units (GPUs)

H

Hyperbolic tangent function

I, J

Implementation, deep learning
Installation, PyTorch

K

Keyword search application

L

Language translation
Laplacian distribution
Leaky ReLU
Lexical ambiguity
Linear function
Linear regression
assumptions
formula
gradient descent algorithm
height and weight
mean, standard deviation and covariance
multiple linear regression model
OLS method
ordinary least square model
prediction errors
predictive modeling
simple linear regression model
specification of
Logistic regression model
Log sigmoid transfer function
Long short-term memory (LSTM) model
Loss function
backward() function
epochs
estimated parameters
final loss value
grad function
hyperparameters
initial value
iteration level
learning rate
linear equation computation
mean square error (MSE)
MSELoss
parameter grid
weight tensor

M

Machine learning
Mean computation
Multidimensional tensor
Multinomial distribution
Multiple linear regression model
Multiprocessing

N

Natural language generation
Natural language processing
applications
five-step approach
Natural language understanding
Network architecture
Neural network (NN)
activation ( see Activation functions)
architecture
data mining
data preparation
definition
design
error functions
functionalities
median, mode and standard deviation
module
Net() function
network architecture
optimization functions
Adadelta
Adagrad
Adam
ASGD
RMSprop algorithm
SGD
SparseAdam
ReLU activation function
set_weight() function
step-by-step approach
structure
tensor differentiation
torch.nn package
n-gram language modeler
Normal distribution
NumPy-based operations

O

Optim module
Optimization function
Adadelta
Adam
backpropagation process
epochs
gradients
loss function
parameters
predicted tensors
regression line
Tensor.backward()
tensor values
torch.no_grad()
training set
validation dataset
Ordinary least square model

P

Predictive text
Probability distribution
autoencoders ( see Autoencoders)
CNN
loss function ( see Loss function)
math operations
model overfitting
dropout rate
hyperparameters
overfitting loss and dropout loss
parameters
predicted values
training accuracy
training dataset
optimization function
RNN
types
weights, dropout rate

Q

Question-and-answering systems

R

Rectified linear unit (ReLu)
Recurrent neural network (RNN)
Adam optimizer
built-in functions
dsets.MINIST() function
embedding layers
hyperparameters
image dataset
LSTM model
memory network
MNIST dataset
predictions
regression problems
cos function
nonlinear cyclical pattern
output layer
test accuracy
test data
time series
weights
Word2vec
Referential ambiguity
Regression learning
RMSprop algorithm

S

Sentiment analysis
Sequential neural network
class Net
functional API
hyperparameters
Keras functions
model architectures
Sigmoid function
Simple linear regression model
Skip gram
SparseAdam
Standard deviation
Statistical distribution
Statistics
Stochastic gradient descent (SGD)
Stochastic variable
Supervised learning
computational graph network
data preparation
forward and backward propagation ( see Neural network (NN))
grad() function
linear regression ( see Linear regression)
logistic regression model
methods
mtcars.csv dataset
nn.backward() method
optimization and gradient computation
training data
Syntactic ambiguity

T

Tensor differentiation
TensorFlow functions
Tensors
arrange function
clamp function
data structure
dimensions
is_storage function
is_tensor function
logarithmic values
LongTensor/index select function
mathematical functions
NumPy functions
1D
split function
transformation functions
2D
unbind function
uniform distribution
Text analysis
Text summarization
Tokenization
Topic modeling
Training data

U, V

Unsupervised learning
Utility function
Utils

W, X, Y, Z

Weight initialization
Word2vec
Word embeddings
context extraction
defined
example
n-gram extractor
vector format