|
absl-py
|
Related to TensorFlow. Abseil Python Common Libraries.
https://pypi.org/project/absl-py/
https://github.com/abseil/abseil-py
|
|
Action potential
|
Neuroscience.
https://en.wikipedia.org/wiki/Action_potential
|
|
Activation function
|
In artificial neural networks, the activation function of a node defines the output of that node given an input or set of inputs.
https://en.wikipedia.org/wiki/Activation_function
https://www.tensorflow.org/api_docs/python/tf/keras/activations
https://www.tensorflow.org/api_docs/python/tf/keras/activations/sigmoid
|
|
AlexNet
|
Es un modelo.
https://en.wikipedia.org/wiki/AlexNet
|
|
Algoritmo de descenso del gradiente
|
See "Gradient Descent algorithm"
https://www.google.com/search?source=hp&ei=wKFSXfKNFaWAjLsPjsW_iAM&q=algoritmo+de+descenso+del+gradiente&oq=algoritmo+de+descenso+del+gradiente&gs_l=psy-ab.3..35i39j0i22i30.2074.2074..3191...0.0..0.117.194.1j1......0....2j1..gws-wiz.LwefKf8B5HI&ved=0ahUKEwjy7YTl4f_jAhUlAGMBHY7iDzEQ4dUDCAU&uact=5
|
|
ANN
|
"Artificial Neural Network".
|
|
artificial intelligence
|
|
|
astor
|
Related to TensorFlow. Es un paquete de Python.
https://pypi.org/project/astor/
|
|
automatic differentiation
|
Related to the computation of the gradient. It can be implemented using C++ operators overloading.
|
|
Axon
|
Neuroscience. It is the output of a neuron. See "Dendrite" and w[k][i][j].
https://en.wikipedia.org/wiki/Axon
|
|
backpropagation
|
Retropropagación. Propagación hacia atrás.
|
|
backpropagation, algoritmo
|
Algoritmo de retropropagación. Algoritmo de propagación hacia atrás.
|
|
bagging
|
Related to "Random forest".
|
|
bias
|
|
|
BigDL
|
Distributed Deep Learning on Apache Spark.
|
|
Bosques Aleatorios
|
Ver "Random forest".
|
|
BrainScript
|
Lenguaje de descripción de modelo de CNTK.
|
|
Caffe
|
"Convolutional Architecture for Fast Feature Embedding". It is a deep learning framework, originally developed at University of California, Berkeley. It is open source, under a BSD license. It is written in C++, with a Python interface. Framework de deep learnig escrito en C++ con un interfaz Python. Soporta muchos tipos diferentes de arquitecturas de deep learning orientadas a la clasificación de imágenes y a la segmentación de imágenes. Soporta CNN, RCNN, LSTM y diseños de redes neuronales conectadas completamente. Caffe soporta librerías de kernel computacional basadas en GPU y CPU tales como NVIDIA cuDNN e Intel MKL. Caffe está siendo usada en proyectos de investigación académica, protitipos startup, e incluso aplicaciones industriales a gran escala en visión, habla y multimedia.
https://en.wikipedia.org/wiki/Caffe_(software)
http://caffe.berkeleyvision.org/
|
|
Caffe, NVIDIA
|
Ver NVCaffe.
http://caffe.berkeleyvision.org/
|
|
Caffe2
|
Framework de deep learning ligero, modular y escalable. Incluye nuevas características como RNNs. Caffe2 es un framework de deep learning diseñado para expresar fácilmente todos los tipos de modelos, por ejemplo CNN, RNN, y más, en un API basado en Python amigable, y ejecutarlos usando un backend de CUDA y C++ altamente eficiente.
https://caffe2.ai/
|
|
CaffeOnSpark
|
Framework de deep learning distribuido creado por Yahoo! usando Spark.
|
|
Cell membrane
|
Neuroscience.
https://en.wikipedia.org/wiki/Cell_membrane
|
|
chain rule
|
Regla de la cadena.
|
|
Chainer
|
Es un framework de deep learning basado en Python apuntando a la flexibilidad. Provee APIs de diferenciación automática basados en el enfoque define-por-ejecución (define-by-run), también conocidas como gráficas computacionales dinámicas, así como APIs de alto nivel orientadas a objetos para construir y entrenar redes neuronales. Soporta CUDA y cuDNN usando CuPy para inferencia y entrenamiento de alto rendimiento.
https://chainer.org/
|
|
Classical conditioning
|
Neuroscience.
|
|
classification algorithms
|
|
|
cluster
|
Conjunto de ordenadores conectados por una red de alta velocidad.
|
|
clustering algorithms
|
Example, "DBSCAN".
|
|
CNN
|
"Convolutional Neural Network".
|
|
CNTK
|
Microsoft "Cognitive Toolkit". The Microsoft Cognitive Toolkit (CNTK) is an open-source toolkit for commercial-grade distributed deep learning. It describes neural networks as a series of computational steps via a directed graph. CNTK allows the user to easily realize and combine popular model types such as feed-forward DNNs, convolutional neural networks (CNNs) and recurrent neural networks (RNNs/LSTMs). CNTK implements stochastic gradient descent (SGD, error backpropagation) learning with automatic differentiation and parallelization across multiple GPUs and servers. CNTK can be included as a library in your Python, C#, or C++ programs, or used as a standalone machine-learning tool through its own model description language (BrainScript). In addition you can use the CNTK model evaluation functionality from your Java programs. CNTK supports 64-bit Linux or 64-bit Windows operating systems. To install you can either choose pre-compiled binary packages, or compile the toolkit from the source provided in GitHub.
https://docs.microsoft.com/en-us/cognitive-toolkit/
https://github.com/Microsoft/CNTK
|
|
Cognitive Toolkit
|
Ver CNTK.
|
|
computer vision
|
|
|
Condicionamiento clásico
|
Neurociencia.
|
|
conductividad
|
Conductivity. Caracteriza cada tipo de sustancia dentro del sistema nervioso. Ver también permeabilidad y permitividad.
|
|
conductivity
|
Characterizes each type of substance within the nervous system. See also permeability and permittivity.
|
|
ConvNet
|
Ver CNN.
|
|
convolution, backward
|
|
|
convolution, forward
|
|
|
CoreNLP
|
Stanford CoreNLP provides a set of human language technology tools. It can give the base forms of words, their parts of speech, whether they are names of companies, people, etc., normalize dates, times, and numeric quantities, mark up the structure of sentences in terms of phrases and syntactic dependencies, indicate which noun phrases refer to the same entities, indicate sentiment, extract particular or open-class relations between entity mentions, get the quotes people said, etc.
https://github.com/stanfordnlp/CoreNLP
https://stanfordnlp.github.io/CoreNLP/
|
|
Cryo-electron tomography
|
See "Electron cryotomography".
|
|
Cryopreservation
|
|
|
cuBLAS
|
The NVIDIA cuBLAS library is a fast GPU-accelerated implementation of the standard basic linear algebra subroutines (BLAS). Using cuBLAS APIs, you can speed up your applications by deploying compute-intensive operations to a single GPU or scale up and distribute work across multi-GPU configurations efficiently. Researchers and scientists use cuBLAS for developing gpu-accelerated algorithms in areas including high performance computing, image analysis and machine learning.
https://developer.nvidia.com/cublas
https://docs.nvidia.com/cuda/cublas/
|
|
CUDA
|
Programming GPUs. Write C-like code that runs directly on the GPU. Optimized APIs: cuBLAS, cuFFT, cuDNN, etc. Es una plataforma de computación paralela y modelo de programación que activa incrementos dramáticos en rendimiento de computación aprovechando el poder de las GPUs NVIDIA.
https://docs.nvidia.com/cuda/
|
|
CUDA Toolkit
|
Develop, Optimize and Deploy GPU-accelerated Apps. he NVIDIA® CUDA® Toolkit provides a development environment for creating high performance GPU-accelerated applications. With the CUDA Toolkit, you can develop, optimize and deploy your applications on GPU-accelerated embedded systems, desktop workstations, enterprise data centers, cloud-based platforms and HPC supercomputers. The toolkit includes GPU-accelerated libraries, debugging and optimization tools, a C/C++ compiler and a runtime library to deploy your application.
https://developer.nvidia.com/cuda-toolkit
|
|
cuDNN
|
NVIDIA "CUDA Deep Neural Network library". Librería de red neuronal profunda NVIDIA CUDA. The NVIDIA CUDA® Deep Neural Network library (cuDNN) is a GPU-accelerated library of primitives for deep neural networks. cuDNN provides highly tuned implementations for standard routines such as forward and backward convolution, pooling, normalization, and activation layers. Deep learning researchers and framework developers worldwide rely on cuDNN for high-performance GPU acceleration. It allows them to focus on training neural networks and developing software applications rather than spending time on low-level GPU performance tuning. cuDNN accelerates widely used deep learning frameworks, including Caffe,Caffe2, Chainer, Keras,MATLAB, MxNet, TensorFlow, and PyTorch.
https://developer.nvidia.com/cudnn
|
|
cuFFT
|
The CUDA Fast Fourier Transform library. The NVIDIA CUDA Fast Fourier Transform library (cuFFT) provides GPU-accelerated FFT implementations that perform up to 10x faster than CPU-only alternatives. Using simple APIs, you can accelerate existing CPU-based FFT implementations in your applications with minimal code changes. As your application grows, you can use cuFFT to scale your image and signal processing applications efficiently across multiple GPUs. It consists of two separate libraries: cuFFT and cuFFTW. The cuFFT library is designed to provide high performance on NVIDIA GPUs. The cuFFTW library is provided as a porting tool to enable users of FFTW to start using NVIDIA GPUs with a minimum amount of effort.
https://docs.nvidia.com/cuda/cufft
https://developer.nvidia.com/cufft
|
|
CuPy
|
CuPy is an open-source matrix library accelerated with NVIDIA CUDA. It also uses CUDA-related libraries including cuBLAS, cuDNN, cuRand, cuSolver, cuSPARSE, cuFFT and NCCL to make full use of the GPU architecture. CuPy's interface is highly compatible with NumPy; in most cases it can be used as a drop-in replacement. All you need to do is just replace numpy with cupy in your Python code. It supports various methods, indexing, data types, broadcasting and more.
https://cupy.chainer.org/
|
|
cuRand
|
|
|
cuSolver
|
|
|
cuSPARSE
|
|
|
Cython
|
Cython is an optimising static compiler for both the Python programming language and the extended Cython programming language (based on Pyrex). It makes writing C extensions for Python as easy as Python itself.
https://cython.org/
|
|
Data Center Accelerator, NVIDIA
|
Acelerador de centro de datos de NVIDIA. Hay varios, por ejemplo: DGX-1, DGX-2.
|
|
data mining
|
Minería de datos. It is the process of discovering patterns in large data sets involving methods of machine learning, statistics and database systems.
|
|
DBSCAN
|
"Density-Based Spatial Clustering of Applications with Noise" (DBSCAN) is a data clustering algorithm.
https://en.wikipedia.org/wiki/DBSCAN
|
|
decision tree
|
Árbol de decisión.
|
|
Deep Cognition Studio (DLS)
|
Es un kit de herramientas de software y plataforma que simplifica el diseño, entrenamiento, testeo y el desarrollo de modelos de IA de deep learning.
https://deepcognition.ai/
|
|
deep learning
|
|
|
Deeplearning4j
|
Eclipse Deeplearning4j is a deep learning programming library written for Java and the Java virtual machine (JVM) and a computing framework with wide support for deep learning algorithms.
|
|
Dendrita
|
Neurociencia. Es una entrada de la neurona. Ver "Axón" y "w[k][i][j]".
https://es.wikipedia.org/wiki/Dendrita
|
|
Dendrite
|
Neuroscience. It is an input of a neuron. A neuron can have more than 1000 dendrites. Each dendrite can be characterized with a weight w[k][i][j].
https://en.wikipedia.org/wiki/Dendrite
|
|
Density-Based Spatial Clustering of Applications with Noise
|
See "DBSCAN".
|
|
design patterns
|
|
|
DFT
|
"Discrete Fourier transform".
https://en.wikipedia.org/wiki/Discrete_Fourier_transform
|
|
DGX Station, NVIDIA
|
Una workstation para equipos de ciencia de datos. NVIDIA DGX Station es la workstation más rápida del mundo para desarrollo de inteligencia artificial avanzada. Este sistema optimizado e integrado completamente permite a tu equipo comenzar más rápido y experimentar con menos esfuerzo con el poder de un centro de datos en tu oficina. 72X más rápida que un servidor de doble CPU.
https://www.nvidia.com/en-us/data-center/dgx-station
|
|
DGX-1, NVIDIA
|
Para investigación en inteligencia artificial. El camino más rápido hacia deep learning. Construir una plataforma para deep learning va mucho más allá de seleccionar un servidor y GPUs. Un compromiso para implementar inteligencia artificial en tu negocio involucra cuidadosamente seleccionar e integrar software complejo con hardware. NVIDIA DGX-1 acelera tu iniciativa con una solución que funciona nada más sacarlo de la caja, por lo que puedes ganar percepciones en horas en lugar de semanas o meses. Red Hat Enterprise Linux en sistemas NVIDIA DGX-1. 140X más rápida que un servidor de una CPU.
https://www.nvidia.com/en-us/data-center/dgx-1
|
|
DGX-2, NVIDIA
|
El sistema de inteligencia artificial más poderoso del mundo para los desafíos de inteligencia artificial más complejos. Experimenta nuevos niveles de escala y velocidad de inteligencia artificial con NVIDIA DGX-2, el primer sistema de 2 petaFLOPS que combina 16 GPUs interconectadas completamente para 10X el rendimiento de deep learning. Es empoderada mediante software NVIDIA DGX y una arquitectura escalable construida sobre NVIDIA NVSwitch, por lo que puedes asumir los desafíos de inteligencia artificial más complejos del mundo. DGX-2, infraestructura esencial de inteligencia artificial Enterprise. 10X más rápida que DGX-1.
https://www.nvidia.com/en-us/data-center/dgx-2
|
|
diferenciación automática
|
Relacionada con el cálculo del gradiente. Puede ser implementada usando sobrecarga de operadores de C++.
|
|
DIGITS, NVIDIA
|
Deep Learning GPU Training System (DIGITS). Sistema de entrenamiento de GPU de deep learning. Pone el poder de deep learning en las manos de ingenieros y científicos de datos. DIGITS puede ser usado para entrenar rápidamente la red neuronal (DNNs) altamente precisa para clasificación de imágenes, segmentación y tareas de detección de objetos.
https://developer.nvidia.com/digits
|
|
DNN
|
"Deep Neural Network". A Deep Neural Network (DNN) is an Artificial Neural Network (ANN) with multiple hidden layers of neurons between the input and output layers.
|
|
DNNL
|
See MKL-DNN.
|
|
Dopamina
|
Neurociencia. Es un neurotransmisor. Cada vez que un neurotransmisor como la dopamina llega a una sinapsis, los circuitos que desencadenan un pensamiento, una motivación o una acción son vía prioritaria en el cerebro. En las adicciones, la dopamina actúa como un neurotransmisor tan potente que las personas, objetos, situaciones y lugares en que se consumió la droga quedan firmemente fijados en la memoria.
https://es.wikipedia.org/wiki/Dopamina
|
|
Dopamine
|
Google repository. Dopamine is a research framework for fast prototyping of reinforcement learning algorithms.
https://github.com/google/dopamine
|
|
Dopamine
|
Neuroscience. It is a neurotransmitter.
https://en.wikipedia.org/wiki/Dopamine
|
|
Electron cryotomography
|
Cryo-electron tomography. VERY HIGH SPATIAL RESOLUTION (~ 1-4 nm).
https://en.wikipedia.org/wiki/Electron_cryotomography
|
|
entrenar red neuronal
|
https://www.tensorflow.org/api_docs/python/tf/keras/Sequential#fit
|
|
entropy, information
|
https://en.wikipedia.org/wiki/Entropy_(information_theory)
|
|
entropía de información
|
Information entropy.
|
|
Error cuadrático medio
|
https://es.wikipedia.org/wiki/Error_cuadr%C3%A1tico_medio
https://www.tensorflow.org/api_docs/python/tf/keras/losses/MeanSquaredError
https://www.tensorflow.org/api_docs/python/tf/keras/Sequential#evaluate
|
|
Fast Fourier Transform
|
FFT.
https://en.wikipedia.org/wiki/Fast_Fourier_transform
|
|
fastText
|
Library for efficient text classification and representation learning. FastText is an open-source, free, lightweight library that allows users to learn text representations and text classifiers. It works on standard, generic hardware. Models can later be reduced in size to even fit on mobile devices.
https://en.wikipedia.org/wiki/FastText
https://fasttext.cc/
https://github.com/facebookresearch/fastText
|
|
feed-forward (red)
|
Prealimentada (alimentada hacia adelante).
|
|
Feedforward Neural Network (FNN)
|
Red neuronal prealimentada (alimentada hacia adelante). A Feedforward Neural Network is an Artificial Neural Network wherein connections between the nodes do not form a cycle. Es una red neuronal en la que las conexiones entre los nodos no forman un ciclo. Ver también RNN.
|
|
FFT
|
"Fast Fourier Transform". Transformada rápida de Fourier.
|
|
FLOPS (flop/s)
|
Floating point operations per second. Operaciones de coma flotante por segundo.
|
|
FNN
|
"Feedforward Neural Network". Red neuronal prealimentada.
|
|
función de activación
|
Ver "Activation function".
|
|
gast
|
TensorFlow. Es un paquete de Python.
https://pypi.org/project/gast/
|
|
Gensim
|
It is a free Python library. Gensim is an open-source library for unsupervised topic modeling and natural language processing, using modern statistical machine learning. It is implemented in Python and Cython.
https://en.wikipedia.org/wiki/Gensim
https://radimrehurek.com/gensim/
|
|
Google AI
|
Google AI is a division of Google dedicated solely to artificial intelligence.
https://ai.google/
|
|
Google Brain (team)
|
https://ai.google/research/teams/brain
|
|
Google I/O
|
Google I/O (or simply I/O) is an annual developer conference held by Google.
https://events.google.com/io
|
|
GoogLeNet
|
Es un modelo.
|
|
gradient
|
https://en.wikipedia.org/wiki/Gradient
|
|
Gradient boosting
|
It is a machine learning technique for regression and classification problems.
https://en.wikipedia.org/wiki/Gradient_boosting
|
|
Gradient Descent algorithm
|
A learning algorithm, training algorithm or optimization algorithm. See also "optimizer".
https://en.wikipedia.org/wiki/Gradient_descent
https://www.tensorflow.org/api_docs/python/tf/train/GradientDescentOptimizer
Set the training algorithm (the optimizer) and the learning rate.
https://www.tensorflow.org/api_docs/python/tf/keras/Sequential#compile
Train the neural network.
https://www.tensorflow.org/api_docs/python/tf/keras/Sequential#fit
|
|
gradiente
|
https://es.wikipedia.org/wiki/Gradiente
|
|
grpcio
|
TensorFlow. Package for gRPC Python.
https://pypi.org/project/grpcio/
|
|
h5py
|
TensorFlow. Es un paquete de Python. The h5py package provides both a high- and low-level interface to the HDF5 library from Python. The low-level interface is intended to be a complete wrapping of the HDF5 API, while the high-level component supports access to HDF5 files, datasets and groups using established Python and NumPy concepts.
https://pypi.org/project/h5py/
http://www.h5py.org/
|
|
Haar training
|
|
|
Hadoop, Apache
|
Java. The Apache Hadoop software library is a framework that allows for the distributed processing of large data sets across clusters of computers using simple programming models.
|
|
HDF
|
"Hierarchical Data Format". It is a set of file formats (HDF4, HDF5) designed to store and organize large amounts of data.
https://en.wikipedia.org/wiki/Hierarchical_Data_Format
https://www.hdfgroup.org/
|
|
HDF5
|
Related to "h5py". File format designed to store and organize large amounts of data. See "HDF".
|
|
HDR
|
High dynamic range.
|
|
Hebb's rule
|
Neuroscience.
|
|
Hebb, regla de
|
Neurociencia. Potenciación a largo plazo (PLP). Regla de Hebb (1949): Si una neurona presináptica A se activa repetidas veces a la vez que la neurona postsináptica B, hay un cambio metabólico que refuerza la eficacia de A excitando B.
|
|
HPC
|
High-performance computing (HPC) is the use of super computers and parallel processing techniques for solving complex computational problems.
|
|
Human Brain Project
|
https://en.wikipedia.org/wiki/Human_Brain_Project
https://www.youtube.com/watch?v=a1XcY-xAvos
https://www.youtube.com/channel/UCd5sWIVavCp4hzp2mFWI2qg
|
|
I/O
|
Ver Google I/O.
|
|
Inception-v3
|
Es un modelo.
|
|
inferencia
|
Cálculo de las salidas de la red neuronal.
Calculate the output of the neural network.
https://www.tensorflow.org/api_docs/python/tf/keras/Sequential#predict
|
|
Integrated Performance Primitives (IPP), Intel
|
It is a multi-threaded software library of functions for multimedia and data processing applications, produced by Intel. It may be used to improve the performance of color conversion, Haar training and DFT functions of the OpenCV library.
https://en.wikipedia.org/wiki/Integrated_Performance_Primitives
https://software.intel.com/en-us/ipp
|
|
IPP
|
See "Integrated Performance Primitives".
|
|
K-means clustering
|
K-medias. It is a method of vector quantization, originally from signal processing, that is popular for cluster analysis in data mining.
https://en.wikipedia.org/wiki/K-means_clustering
|
|
K-medias
|
K-means clustering. Método utilizado en minería de datos.
|
|
Keras
|
Keras is a high-level API to build and train deep learning models.
https://keras.io/
https://www.tensorflow.org/api_docs/python/tf/keras
|
|
Keras-Applications
|
TensorFlow. Es un paquete de Python.
https://pypi.org/project/Keras-Applications/
https://keras.io/applications/
|
|
Keras-Preprocessing
|
TensorFlow. Es un paquete de Python.
https://pypi.org/project/Keras-Preprocessing/
https://keras.io/
|
|
Learning
|
Computing.
|
|
Learning
|
Neuroscience.
|
|
learning algorithm
|
Learning algorithm, training algorithm or optimization algorithm.
https://www.google.com/search?source=hp&ei=9LFCXcWeMeyOlwT6_Z2YAw&q=%22neural+network%22+%22learning+algorithms%22+site%3Aedu&oq=%22neural+network%22+%22learning+algorithms%22+site%3Aedu&gs_l=psy-ab.3...1471.1471..4314...0.0..0.110.174.1j1......0....2j1..gws-wiz.....0.5qTRLms_d90&ved=0ahUKEwjF6Jb_ruHjAhVsx4UKHfp-BzMQ4dUDCAU&uact=5
https://www.google.com/search?source=hp&ei=WLJCXfnDFI-TlwTl7KU4&q=%22neural+network%22+%22learning+algorithms%22+site%3Aac.uk&oq=%22neural+network%22+%22learning+algorithms%22+site%3Aac.uk&gs_l=psy-ab.3...1525.1525..2746...0.0..0.116.175.1j1......0....2j1..gws-wiz.....0.49O3UCL1o2s&ved=0ahUKEwj5z9Gur-HjAhWPyYUKHWV2CQcQ4dUDCAU&uact=5
|
|
Levenberg–Marquardt algorithm
|
https://en.wikipedia.org/wiki/Levenberg%E2%80%93Marquardt_algorithm
|
|
LSTM network
|
"Long Short-Term Memory" network. It is an artificial Recurrent Neural Network (RNN) architecture used in the field of deep learning.
|
|
LSTM, red
|
"Long Short-Term Memory". Es un tipo especial de red neuronal recurrente (RNN) capaz de aprender dependencias long-term (largo plazo).
|
|
machine learning
|
|
|
map nervous system
|
Extract the neural network from a dead person or from a dead animal using tomography images of high resolution.
|
|
mapear sistema nervioso
|
Consiste en extraer la información de la red neuronal de una persona muerta o de un animal muerto usando imágenes de tomografía de muy alta resolución. Todavía no se puede hacer (que yo sepa). Tal vez no se pueda hacer en muchas décadas. El tiempo de escaneo será inaceptable. Hay que cortar y destruir el sistema nervioso para poder extraer la información. No basta con tener resolución espacial y resolución de contraste suficientes. El efecto de la congelación o del método que se use para fijar las neuronas se podrá deshacer mediante software. La fijación de las neuronas (congelación u otro método) se tendrá que hacer antes de que las neuronas se degraden más allá de lo que permita el método para extraer la información de la red neuronal. Ver "Cryopreservation".
|
|
Markdown
|
TensorFlow. Es un paquete de Python.
https://pypi.org/project/Markdown/
https://python-markdown.github.io/
|
|
MATLAB
|
It is a multi-paradigm numerical computing environment and proprietary programming language developed by MathWorks.
https://en.wikipedia.org/wiki/MATLAB
|
|
MATLAB Deep Learning Toolbox
|
https://www.mathworks.com/products/deep-learning.html
https://www.mathworks.com/help/deeplearning/ug/choose-a-multilayer-neural-network-training-function.html
|
|
MATLAB Deep Network Designer
|
https://www.mathworks.com/help/deeplearning/ref/deepnetworkdesigner-app.html
|
|
Matplotlib
|
Matplotlib is a plotting library for the Python programming language and its numerical mathematics extension NumPy.
https://en.wikipedia.org/wiki/Matplotlib
https://matplotlib.org/
|
|
Mean squared error
|
https://en.wikipedia.org/wiki/Mean_squared_error
https://www.tensorflow.org/api_docs/python/tf/keras/losses/MeanSquaredError
Calculate the "mean square error" (mse).
https://www.tensorflow.org/api_docs/python/tf/keras/Sequential#evaluate
|
|
membrana celular
|
Referido a neurona animal. La forma de la membrana celular de una dendrita y otra información determinan el valor del peso w[k][i][j] de esa dendrita.
|
|
menoh
|
Fast DNN inference library with multiple programming language support.
https://github.com/pfnet-research/menoh
|
|
mente
|
Red neuronal.
|
|
mind
|
Neural network.
|
|
minería de datos
|
Data mining. Intenta descubrir patrones en grandes volúmenes de de datos.
|
|
MKL, Intel
|
Intel "Math Kernel Library". Librería de núcleo matemático de Intel. Intel Math Kernel Library (Intel MKL) is a library of optimized math routines for science, engineering, and financial applications. Core math functions include BLAS, LAPACK, ScaLAPACK, sparse solvers, fast Fourier transforms, and vector math. Es una librería de rutinas matemáticas optimizadas para ciencia, ingeniería y aplicaciones financieras. Las funciones matemáticas núcleo incluyen BLAS, LAPACK, ScaLAPACK, solucionadores dispersos, transformada de Fourier rápida y matemáticas de vector. Están optimizadas para procesadores Intel. Esta librería está disponible para Windows, Linux y macOS. Uses industry-standard C and Fortran APIs for compatibility with popular BLAS, LAPACK, and FFTW functions—no code changes required. Improved performance of 1D/3D FFT (fast Fourier transform).
https://en.wikipedia.org/wiki/Math_Kernel_Library
https://software.intel.com/mkl
|
|
MKL-DNN, Intel
|
Intel "Math Kernel Library for Deep Neural Networks". Librería de núcleo de matemáticas de Intel para redes neuronales profundas. Starting with version 1.1 the library is renamed to DNNL. Deep Neural Network Library (DNNL) is an open-source performance library for deep learning applications. The library includes basic building blocks for neural networks optimized for Intel Architecture Processors and Intel Processor Graphics. DNNL is intended for deep learning applications and framework developers interested in improving application performance on Intel CPUs and GPUs. Deep learning practitioners should use one of the applications enabled with DNNL.
https://github.com/intel/mkl-dnn
|
|
ML
|
"Machine Learning".
|
|
MLP
|
"Multilayer Perceptron".
https://en.wikipedia.org/wiki/Multilayer_perceptron
|
|
mlpack
|
mlpack is a machine learning software library for C++.
https://en.wikipedia.org/wiki/Mlpack
https://mlpack.org/
|
|
Model
|
Neural network.
|
|
Modelo
|
Red neuronal.
|
|
MXNet
|
Es un framework de deep learning que te permite mezclar los sabores de programación simbólica y programación imperativa para maximizar la eficiencia y la productividad. La libería es portable y ligera, y se escala a múltiples GPUs y múltiples máquinas.
https://mxnet.apache.org/
|
|
Máquina de vectores de soporte
|
Support-vector machine.
|
|
NCCL
|
Related to "CuPy".
|
|
Nernst equation
|
Neuroscience.
https://en.wikipedia.org/wiki/Nernst_equation
|
|
Nervous system
|
Neuroscience.
https://en.wikipedia.org/wiki/Nervous_system
|
|
neural network
|
Example of simple neural network in Python with TensorFlow.
https://xformulas.net/source_code/python/simple_neural_network.py.php
|
|
neural network
|
|
|
Neuron
|
Neuroscience. A typical neuron have several dendrites (inputs), a cell nucleus, and an axon (output).
https://en.wikipedia.org/wiki/Neuron
|
|
neurona
|
Neurociencia. Una neurona típica tiene varias dendritas (entradas), un núcleo celular, y un axón (salida).
https://es.wikipedia.org/wiki/Neurona
|
|
Neuroscience
|
https://en.wikipedia.org/wiki/Neuroscience
|
|
neurotransmisor
|
Neurociencia.
https://es.wikipedia.org/wiki/Neurotransmisor
|
|
neurotransmitter
|
Neuroscience.
https://en.wikipedia.org/wiki/Neurotransmitter
|
|
nGraph
|
Open source C++ library, compiler and runtime for Deep Learning. nGraph Compiler aims to accelerate developing AI workloads using any deep learning framework and deploying to a variety of hardware targets.
https://github.com/NervanaSystems/ngraph
https://www.ngraph.ai/
|
|
NLP
|
"Natural Language Processing".
|
|
NLTK
|
"Natural Language Toolkit".
|
|
NNP-I
|
Intel® Nervana™ Neural Network Processor for Inference. Related to "nGraph".
|
|
NNP-L
|
Intel® Nervana™ Neural Network Processor for Learning. Related to "nGraph".
|
|
NumPy
|
NumPy is a scientific computing package for Python.
https://www.numpy.org/
import numpy as np
xx = np.array([[5.0, 1.0, 0.0, 3.0]])
print(xx.shape)
|
|
NVCaffe
|
NVIDIA Caffe. Es un fork mantenido por NVIDIA de Caffe ajustado para GPUs NVIDIA, particularmente en configuraciones multi GPU. Incluye soporte de multiprecisión así como otras características mejoradas de NVIDIA y ofrece rendimiento especialmente ajustado para NVIDIA DGX-2, DGX-1 y DGX Station.
http://docs.nvidia.com/deeplearning/dgx/caffe-release-notes/index.html
http://caffe.berkeleyvision.org/
|
|
NVSwitch, NVIDIA
|
Se usa en DGX-2.
|
|
ONNX
|
"Open Neural Network Exchange". ONNX is an open format to represent deep learning models. With ONNX, AI developers can more easily move models between state-of-the-art tools and choose the combination that is best for them. ONNX is developed and supported by a community of partners. Enabling interoperability makes it possible to get great ideas into production faster. ONNX enables models to be trained in one framework and transferred to another for inference. ONNX models are currently supported in Caffe2, Microsoft Cognitive Toolkit, MXNet, and PyTorch, and there are connectors for many other common frameworks and libraries. Es un formato para modelos de deep learning que permite interoperabilidad entre diferentes frameworks de IA. ONNX es una representación de modelo compartida para interoperabilidad framework y optimización compartida. ONNX permite a los desarrolladores mover modelos entre frameworks tales como CNTK, Caffe2, MXNet y PyTorch.
https://onnx.ai/
|
|
ONNX Runtime
|
ONNX Runtime is a performance-focused complete scoring engine for Open Neural Network Exchange (ONNX) models, with an open extensible architecture to continually address the latest developments in AI and Deep Learning. ONNX Runtime stays up to date with the ONNX standard with complete implementation of all ONNX operators, and supports all ONNX releases (1.2+) with both future and backwards compatibility. Please refer to this page for ONNX opset compatibility details.
https://github.com/microsoft/onnxruntime
|
|
OpenCL
|
Similar to CUDA, but runs on anything. Usually slower on NVIDIA hardware.
https://en.wikipedia.org/wiki/OpenCL
|
|
OpenCV
|
OpenCV (Open source computer vision) is a library of programming functions mainly aimed at real-time computer vision.
https://en.wikipedia.org/wiki/OpenCV
https://opencv.org/
|
|
OpenNI
|
"Open Natural Interaction". The OpenNI Framework contains a set of open source APIs that provide support for natural interaction with devices via methods such as voice command recognition, hand gestures, and body motion tracking.
|
|
OpenVino toolkit
|
Develop applications and solutions that emulate human vision with the Intel® Distribution of OpenVINO™ toolkit. Based on convolutional neural networks (CNN), the toolkit extends workloads across Intel® hardware (including accelerators) and maximizes performance.
https://software.intel.com/en-us/openvino-toolkit
|
|
OpenVX
|
OpenVX is an open, royalty-free standard for cross platform acceleration of computer vision applications.
https://en.wikipedia.org/wiki/OpenVX
|
|
Optical coherence tomography
|
High spatial resolution (< 1 um). See also "Cryo-electron tomography".
https://en.wikipedia.org/wiki/Optical_coherence_tomography
|
|
optimization algorithm
|
Learning algorithm, training algorithm or optimization algorithm.
https://www.google.com/search?source=hp&ei=U7NCXbyVHoacaca7kxA&q=%22neural+network%22+%22optimization+algorithms%22+site%3Aedu&oq=%22neural+network%22+%22optimization+algorithms%22+site%3Aedu&gs_l=psy-ab.3...1462.1462..2038...0.0..0.106.170.1j1......0....2j1..gws-wiz.....0.xKJVh5eBOhM&ved=0ahUKEwj8irOmsOHjAhUGThoKHcbdBAIQ4dUDCAU&uact=5
https://www.google.com/search?source=hp&ei=c7NCXbzcK8yua-7hg8AB&q=%22neural+network%22+%22optimization+algorithms%22+site%3Aac.uk&oq=%22neural+network%22+%22optimization+algorithms%22+site%3Aac.uk&gs_l=psy-ab.3...1195.1195..2858...0.0..0.100.175.1j1......0....2j1..gws-wiz.....0.sg2H4zd0US4&ved=0ahUKEwj84eG1sOHjAhVM1xoKHe7wABgQ4dUDCAU&uact=5
|
|
optimizer
|
TensorFlow.
https://www.tensorflow.org/api_docs/python/tf/train/GradientDescentOptimizer
Set the training algorithm (the optimizer) and the learning rate.
https://www.tensorflow.org/api_docs/python/tf/keras/Sequential#compile
|
|
overfitting
|
Sobreajuste.
|
|
PaddlePaddle
|
Proporciona un interfaz intuitivo y flexible para cargar datos y especificar estructuras de modelos. Soporta CNN, RNN, variantes múltiples y configura modelos profundos complicados fácilmente. Proporciona operaciones optimizadas extremadamente, reciclaje de memoria y comunicación de red. PaddlePaddle hace fácil escalar recursos de computación heterogeneos y almacenamiento para acelerar el proceso de entrenamiento.
http://www.paddlepaddle.org/
|
|
pandas
|
|
|
pasta
|
TensorFlow. Es un paquete de Python.
https://github.com/google/pasta
https://pypi.org/project/google-pasta/
|
|
permeabilidad
|
Permeability. Caracteriza cada tipo de sustancia dentro del sistema nervioso. Ver también conductividad y permitividad.
|
|
permeability
|
Characterizes each type of substance within the nervous system. See also conductivity and permittivity.
https://en.wikipedia.org/wiki/Permeability_(electromagnetism)
|
|
permitividad
|
Permittivity. Caracteriza cada tipo de sustancia dentro del sistema nervioso. Ver también conductividad y permeabilidad.
|
|
permittivity
|
Characterizes each type of substance within the nervous system. See also conductivity and permeability.
https://en.wikipedia.org/wiki/Permittivity
|
|
peta
|
10^15
|
|
Placer
|
Neurociencia.
|
|
PlaidML
|
|
|
Pleasure
|
Neuroscience.
http://thebrain.mcgill.ca/flash/d/d_03/d_03_cr/d_03_cr_que/d_03_cr_que.html
|
|
pooling
|
|
|
Potencial de acción
|
Neurociencia. Los "potenciales de acción" se propagan extremadamente despacio, a unos 10 m/s.
https://es.wikipedia.org/wiki/Potencial_de_acci%C3%B3n
"Las velocidades de conducción del potencial de acción varían, si bien 10 m/s es una velocidad típica. Debemos recordar que desde el inicio hasta el final la duración del potencial de acción es de unos 2 ms. A partir de este dato, podemos calcular la longitud de la membrana que está implicada en el potencial de acción en cualquier momento:
10 m/s x 2 x 10^-3 s = 2 x 10^-2 m.
Así pues, un potencial de acción que se desplaza a 10 m/s abarca 2 cm del axón."
Transcripción de fragmento del libro "NEUROCIENCIA. La exploración del cerebro", 3ª edición, de Mark F. Bear, Barry Connors y Michael Paradiso. Capítulo 4, página 94.
|
|
PowerAI Enterprise, IBM
|
Plataforma de inteligencia artificial de IBM.
https://www.ibm.com/us-en/marketplace/deep-learning-platform
|
|
propagation constant
|
|
|
protobuf
|
TensorFlow. Es un paquete de Python.
|
|
Protocol Buffers
|
|
|
PyPI
|
The Python Package Index (PyPI) is a repository of software for the Python programming language. PyPI helps you find and install software developed and shared by the Python community. Learn about installing packages. Package authors use PyPI to distribute their software.
https://pypi.org/
|
|
Pyro, Uber
|
Software para programación probabilística.
|
|
Python
|
Python is an interpreted, high-level, general-purpose programming language.
https://en.wikipedia.org/wiki/Python_(programming_language)
|
|
PyTorch
|
An open source machine learning framework that accelerates the path from research prototyping to production deployment. Es una librería de machine learning para Python basada en Torch usada para aplicaciones tales como procesado de lenguaje natural. Está desarrollada principalmente por el grupo de investigación de inteligencia artificial de Facebook. Proporciona 2 características de alto nivel: computación "Tensor" (como NumPy) con aceleración GPU fuerte, y redes neuronales profundas construidas sobre un sistema autodiff basado en cinta. PyTorch soporta varios tipos de tensores. PyTorch es un framework computacional de tensores acelerado por GPU con front end Python. Funcionalmente puede ser fácilmente extendido con librerías Python comunes tales como NumPy, SciPy y Cython. La diferenciación automática es realizada con un sistema basado en cinta en ambos un funcional y nivel de capa de red neuronal. Esta funcionalidad trae un alto nivel de flexibilidad y velocidad como un framework deep learning y proporciona funcionalidad tipo NumPy acelerada. PyTorch es una plataforma de aprendizaje profundo de fuente abierta que provee un camino sin problemas desde prototipado de investigación hasta desarrollo de producción.
https://pytorch.org/
|
|
PyTorch Tensor
|
Similares a los arrays NumPy, con el añadido de que los tensores pueden también ser usados sobre GPUs que soporten CUDA.
|
|
R
|
|
|
R-CNN
|
La "R" se refiere a región. Para detección de objetos. Ver CNN.
|
|
Random forest
|
Bosques Aleatorios.
|
|
RCNN
|
Ver RCNN.
|
|
red neuronal
|
|
|
Red neuronal prealimentada
|
Feedforward Neural Network (FNN).
|
|
Red neuronal recurrente
|
Recurrent Neural Network (RNN).
|
|
Regla de Hebb
|
Neurociencia. Hebb's rule. Potenciación a largo plazo (PLP). Regla de Hebb (1949): Si una neurona presináptica A se activa repetidas veces a la vez que la neurona postsináptica B, hay un cambio metabólico que refuerza la eficacia de A excitando B.
|
|
regression algorithms
|
|
|
ResNet-101
|
Es un modelo.
|
|
ResNet-50
|
Es un modelo.
|
|
resolución de contraste (de tomografía)
|
|
|
resolución espacial (de tomografía)
|
Este parámetro determina brutalmente cual será el curso de la humanidad. La mente (red neuronal) de una persona se puede almacenar indefinidamente en una imagen tridimensional de tomografía de resoluciones espacial y de color o contraste suficientes. El volumen de las neuronas, la forma de las neuronas, la posición de las neuronas, la forma de las sustancias diferentes dentro de las neuronas, ... determinan los datos numéricos (ver "w[k][i][j]") con los que se puede recrear cada neurona dentro de un sistema hardware software.
|
|
Reward circuit
|
Neuroscience.
|
|
Reward system
|
Neuroscience.
https://en.wikipedia.org/wiki/Reward_system
|
|
RNN
|
"Recurrent Neural Network". Red neuronal recurrente. Es una clase de red neuronal artificial donde las conexiones entre nodos forman un gráfico dirigido a lo largo de una secuencia. Esto permite exhibir un comportamiento dinámico temporal para una secuencia de tiempo. A diferencia de las FNNs, las RNNs pueden usar su estado interno (memoria) para procesar secuencias de entradas.
https://en.wikipedia.org/wiki/Recurrent_neural_network
|
|
Scikit-learn
|
It is a free software machine learning library for the Python programming language. Simple and efficient tools for data mining and data analysis. Built on NumPy, SciPy, and matplotlib.
https://scikit-learn.org/
|
|
SciPy
|
It is a free and open-source Python library used for scientific computing and technical computing. SciPy contains modules for optimization, linear algebra, integration, interpolation, special functions, FFT, signal and image processing, ODE solvers and other tasks common in science and engineering. SciPy builds on the NumPy array object and is part of the NumPy stack which includes tools like Matplotlib, pandas and SymPy, and an expanding set of scientific computing libraries.
https://en.wikipedia.org/wiki/SciPy
https://scipy.org/scipylib/
|
|
SGD
|
"Stochastic Gradient Descent". Descenso de gradiente estocástico.
|
|
Sigmoid function
|
https://en.wikipedia.org/wiki/Sigmoid_function
https://www.tensorflow.org/api_docs/python/tf/keras/activations/sigmoid
|
|
Sigmoide, función
|
https://es.wikipedia.org/wiki/Funci%C3%B3n_sigmoide
|
|
Sinapsis
|
Neurociencia.
https://es.wikipedia.org/wiki/Sinapsis
|
|
Sistema de recompensa
|
Neurociencia. Ver "Reward system".
|
|
Sistema Nervioso animal
|
Neurociencia.
Si tratamos de buscar semejanzas de una red neuronal artificial con el sistema nervioso animal, cada peso en el sistema nervioso animal (w[k][i][j]) sería algo parecido a las características de la sinapsis y las neuritas presináptica y postsináptica.
La repetición para evitar el dolor y maximizar la "satisfacción" (placer) a lo largo del tiempo, es lo que provoca el aprendizaje en el sistema nervioso animal. En las redes neuronales el aprendizaje también se consigue mediante la repetición (los pesos w[k][i][j] cambian solamente un poco en cada iteración). El mayor o menor tránsito de potenciales de acción cambia las propiedades eléctricas de las sinapsis y las neuronas.
Cuanto mejor son satisfechas las necesidades (menos dolor, más "satisfacción"), más dopamina se libera, que provoca que aumente el tamaño y la efectividad de los circuitos neuronales (pensamientos) que se encuentren activos, es decir, aumentan los pesos w[k][i][j] de las conexiones de esos circuitos.
El objetivo del sistema nervioso parece que debería ser satisfacer lo mejor posible las necesidades a lo largo del tiempo. Esto está fuertemente relacionado con la Teoría de la Evolución de Charles Darwin, y parece claro que es una consecuencia necesaria de dicha teoría.
Sigmund Freud también se refiere al placer y dice algo parecido (entre otras cosas).
En general el objetivo del sistema nervioso es maximizar el placer (menos dolor, más "satisfacción") desde el presente hasta un tiempo futuro que depende mucho de lo aprendido (incluidas las creencias religiosas).
Si pudiéramos medir en cada instante con un número como de bien están siendo satisfechas las necesidades, el objetivo del sistema nervioso debería ser maximizar la integral de ese número desde el instante de tiempo presente hasta el tiempo en el que vamos a morir, o incluso hasta el infinito, si es lo que dice tu religión y estás absolutamente convencido de que es cierto lo que dice tu religión.
Ver "Regla de Hebb", "Reward system" y "dopamina".
|
|
six
|
TensorFlow. Es un paquete de Python.
|
|
snippet
|
[Referred to OpenCV]
|
|
sobreajuste
|
Overfitting. Es el efecto de sobreentrenar con un algoritmo de aprendizaje.
|
|
Support-vector machines
|
Máquinas de vectores de soporte. These are supervised learning models with associated learning algorithms that analyze data used for classification and regression analysis.
https://en.wikipedia.org/wiki/Support-vector_machine
|
|
SVM
|
"Support-vector machine".
|
|
Synapse
|
Neuroscience.
https://en.wikipedia.org/wiki/Synapse
|
|
TBB
|
See "Threading Building Blocks".
|
|
Tensor
|
Array multidimensional.
|
|
tensorboard
|
TensorFlow. Es un paquete de Python.
|
|
TensorFlow
|
TensorFlow is an end-to-end open source platform for machine learning. It has a comprehensive, flexible ecosystem of tools, libraries and community resources that lets researchers push the state-of-the-art in ML and developers easily build and deploy ML powered applications. Build and train ML models easily using intuitive high-level APIs like Keras with eager execution, which makes for immediate model iteration and easy debugging. Easily train and deploy models in the cloud, on-prem, in the browser, or on-device no matter what language you use. TensorFlow is an open source software library for high performance numerical computation. Its flexible architecture allows easy deployment of computation across a variety of platforms (CPUs, GPUs, TPUs), and from desktops to clusters of servers to mobile and edge devices. Originally developed by researchers and engineers from the Google Brain team within Google’s AI organization, it comes with strong support for machine learning and deep learning and the flexible numerical computation core is used across many other scientific domains.
https://www.tensorflow.org/
https://pypi.org/project/tensorflow/
|
|
tensorflow-estimator
|
TensorFlow. TensorFlow Estimator is a high-level API that encapsulates model training, evaluation, prediction, and exporting.
https://pypi.org/project/tensorflow-estimator/
|
|
TensorRT Inference Server
|
Servidor de inferencia de TensorRT. Provee una solución de inferencia de centro de datos optimizada para GPUs NVIDIA. Maximiza la utilización de inferencia y rendimiento sobre GPUs vía un punto final de gRPC o HTTP, permitiendo a clientes remotos pedir inferencia para cualquier modelo que esté siendo manejado por el servidor, así como suministrando métricas en tiempo real sobre latencia y peticiones. El servidor de inferencia TensorRT provee una solución de inferenciación de nube optimizada para GPUs NVIDIA.
https://docs.nvidia.com/deeplearning/sdk/
|
|
TensorRT, NVIDIA
|
NVIDIA TensorRT™ is a platform for high-performance deep learning inference. It includes a deep learning inference optimizer and runtime that delivers low latency and high-throughput for deep learning inference applications. TensorRT-based applications perform up to 40x faster than CPU-only platforms during inference. With TensorRT, you can optimize neural network models trained in all major frameworks, calibrate for lower precision with high accuracy, and finally deploy to hyperscale data centers, embedded, or automotive product platforms.
https://developer.nvidia.com/tensorrt
http://docs.nvidia.com/deeplearning/sdk/
|
|
tera
|
10^12
|
|
termcolor
|
TensorFlow. Es un paquete de Python.
https://pypi.org/project/termcolor/
|
|
TF
|
See "TensorFlow".
|
|
Theano
|
Theano es una librería Python que te permite definir, optimizar, y evaluar expresiones matemáticas que involucran arrays multidimensionales eficientemente. Theano ha estado alimentando investigaciones científicas intensivas computacionalmente de gran escala desde 2007. Pero es también bastante accesible para ser usado en el aula.
https://en.wikipedia.org/wiki/Theano_(software)
https://github.com/Theano/
http://deeplearning.net/software/theano
|
|
Threading Building Blocks, Intel
|
TBB. Intel® Threading Building Blocks (Intel® TBB) is a widely used C++ library for shared memory parallel programming and heterogeneous computing (intra-node distributed memory programming). The library provides a wide range of features for parallel programming that include: Generic parallel algorithms, Concurrent containers, A scalable memory allocator, Work-stealing task scheduler, Low-level synchronization primitives. It is used inside OpenCV for parallel code snippets. Using this will make sure that the OpenCV library will take advantage of all the cores you have in your system's CPU.
https://software.intel.com/en-us/tbb
https://github.com/intel/tbb
|
|
Tomografía
|
https://es.wikipedia.org/wiki/Tomograf%C3%ADa
https://elpais.com/sociedad/2013/06/20/actualidad/1371742600_459472.html
|
|
Tomography
|
https://en.wikipedia.org/wiki/Tomography
https://www.youtube.com/watch?v=ldXEuUVkDuw
https://www.youtube.com/watch?v=aoUgQtaotg4
https://www.youtube.com/watch?v=gv6Gxbp8Rvs
|
|
Torch
|
Torch is a scientific computing framework with wide support for machine learning algorithms that puts GPUs first. It is easy to use and efficient, thanks to an easy and fast scripting language, LuaJIT, and an underlying C/CUDA implementation.
https://en.wikipedia.org/wiki/Torch_(machine_learning)
http://torch.ch/
|
|
TPU
|
"Tensor Processing Unit". A tensor processing unit (TPU) is an AI accelerator application-specific integrated circuit (ASIC) developed by Google specifically for neural network machine learning.
https://en.wikipedia.org/wiki/Tensor_processing_unit
https://cloud.google.com/blog/products/gcp/an-in-depth-look-at-googles-first-tensor-processing-unit-tpu
https://cloud.google.com/tpu/
|
|
train neural network
|
See "learning".
Train the neural network.
https://www.tensorflow.org/api_docs/python/tf/keras/Sequential#fit
|
|
training algorithm
|
Learning algorithm, training algorithm or optimization algorithm.
https://www.google.com/search?source=hp&ei=jLJCXYHkKY-WlwSpmYjoDQ&q=%22neural+network%22+%22training+algorithms%22+site%3Aedu&oq=%22neural+network%22+%22training+algorithms%22+site%3Aedu&gs_l=psy-ab.3...1373.1373..1962...0.0..0.99.167.2......0....2j1..gws-wiz.....0.9jpwS9FuxFQ&ved=0ahUKEwiB2szHr-HjAhUPy4UKHakMAt0Q4dUDCAU&uact=5
https://www.google.com/search?source=hp&ei=v7JCXd9wkJuXBOjfpYAM&q=%22neural+network%22+%22training+algorithms%22+site%3Aac.uk&oq=%22neural+network%22+%22training+algorithms%22+site%3Aac.uk&gs_l=psy-ab.3...1555.1555..3421...0.0..0.106.170.1j1......0....2j1..gws-wiz.....0.dObZ2OrNb4I&ved=0ahUKEwifzMzfr-HjAhWQzYUKHehvCcAQ4dUDCAU&uact=5
|
|
training function
|
See "training algorithm".
|
|
VGG-16
|
Es un modelo.
|
|
VGG-19
|
Es un modelo.
|
|
w[k][i][j]
|
Peso entre la neurona i de la capa (k-1) y la neurona j de la capa k.
|
|
Watson
|
https://en.wikipedia.org/wiki/Watson_(computer)
https://elpais.com/diario/2011/02/18/radiotv/1297983603_850215.html
https://www.youtube.com/user/IBMWatsonSolutions
https://www.ibm.com/watson
|
|
wave propagation
|
https://en.wikipedia.org/wiki/Wave_propagation
|
|
Werkzeug
|
TensorFlow. Werkzeug is a comprehensive WSGI web application library. It began as a simple collection of various utilities for WSGI applications and has become one of the most advanced WSGI utility libraries.
https://pypi.org/project/Werkzeug/
https://palletsprojects.com/p/werkzeug/
|
|
Word2vec
|
It is a group of related models that are used to produce word embeddings.
https://en.wikipedia.org/wiki/Word2vec
|
|
wrapt
|
TensorFlow. The aim of the wrapt module is to provide a transparent object proxy for Python, which can be used as the basis for the construction of function wrappers and decorator functions.
https://pypi.org/project/wrapt/
https://github.com/GrahamDumpleton/wrapt
|