Presentacion final Machine Learning LSTM Networks.
Usefull Machine Learning Software. (for a change)

Warnings:
There are some attempts at humor here.
----
# Recurrent Neural Networks (RNN)

----
## Long-Short Term Memory (LSTM) Network
### Recurrent Neural Networks (RNN)
[](http://colah.github.io/posts/2015-08-Understanding-LSTMs/)
- Image caption
- Secuence classfication
- Translation
- Named entity recognition
_
====
### Recurrent Neural Networks have loops
Humans don’t start their thinking from scratch every second.
[](http://colah.github.io/posts/2015-08-Understanding-LSTMs/)
Recurrent neural networks address this issue. They are networks with loops in them, allowing information to persist.
_
====
### An unrolled recurrent neural network.
A recurrent neural network can be thought of as multiple copies of the same network, each passing a message to a successor.
[](http://colah.github.io/posts/2015-08-Understanding-LSTMs/)
_
====
### The Problem of Long-Term Dependencies
[](http://colah.github.io/posts/2015-08-Understanding-LSTMs/)
Predict the next word:
- the clouds are in the **sky**,
- I grew up in France… I speak fluent **French**.
_
====
## LSTM Networks
Long Short Term Memory networks – usually just called “LSTMs” – are a special kind of RNN, capable of learning long-term dependencies.
_
====
### The repeating module in a standard RNN contains a single layer.
[](http://colah.github.io/posts/2015-08-Understanding-LSTMs/)
_
====
### The repeating module in an LSTM contains four interacting layers.
[](http://colah.github.io/posts/2015-08-Understanding-LSTMs/)
[](http://colah.github.io/posts/2015-08-Understanding-LSTMs/)
_
====
### More Neurons,Layers: More Deep Learning.
[](http://www.slideshare.net/holbertonschool/deep-learning-keynote-1-by-louis-monier)
----
**Node.js**
Es un entorno de ejecución para JavaScript basado en el motor V8 de Google y actualmente cuenta con el ecosistema de librerias de código abierto mas grande del mundo llamada NPM. Es un entorno de ejecución orientado a eventos.

_
====
**Qué es y para que sirve Node.js?**
Node.js es un entorno de ejecución de JavaScript de lado del servidor, el cual permite la creación de aplicaciones de red altamente escalables.

_
====
**Qué es y para que sirve Node.js?**
Se basa en Programación Orientada a Eventos Asíncronos. Node.js cambia la manera de como conectarse a un servidor, ya que cada conexión dispara un evento especifico dentro del motor y se afirma que puede soportar decenas de miles de conexiones concurrentes sin bloqueos.

_
====
**JavaScript en el servidor**
- Lenguaje de programación potente
- Posee un excelente modelo de eventos asincrónicos
- Curva de aprendizaje reducida debido al previo conocimiento de muchos desarrolladores

_
====
**Ventajas**
- Gran capacidad de tolerancia a la cantidad de peticiones simultaneas
- Existencia de gran cantidad de módulos para casi todas las necesidades que se tengan gracias a NPM
- Menor costo de infraestructura
- Comunidad creciente dispuesta a resolver dudas y mostrar nuevas maneras de uso

----
## Demo [IAMDinosaur](https://www.youtube.com/watch?v=P7XHzqZjXQs)
[](https://github.com/ivanseidel/IAMDinosaur)
_
====
### Chrome/Chromium offline script
[](http://caza.la/synaptic/#/)
- [Chromium](https://chromium.googlesource.com/chromium/src.git/+/master/components/neterror/resources/offline.js)
- [Code Google.](https://code.google.com/p/chromium/codesearch#chromium/src/components/neterror/resources/offline.js&q=offline.js&sq=package:chromium&type=cs)
_
====
### Synaptic
[](http://caza.la/synaptic/#/)
_
====
### RobotJs
[](https://github.com/octalmage/robotjs)
_
====
### Sensors
[](http://www.slideshare.net/NodejsFoundation/from-pterodactyls-and-cactus-to-artificial-intelligence-ivan-seidel-gomes-tenda-digital-66868648)
_
====
### Actuators
[](http://www.slideshare.net/NodejsFoundation/from-pterodactyls-and-cactus-to-artificial-intelligence-ivan-seidel-gomes-tenda-digital-66868648)
_
====
### Neural Network
The distance affects the input
[](http://www.slideshare.net/NodejsFoundation/from-pterodactyls-and-cactus-to-artificial-intelligence-ivan-seidel-gomes-tenda-digital-66868648)
_
====
[](http://www.slideshare.net/NodejsFoundation/from-pterodactyls-and-cactus-to-artificial-intelligence-ivan-seidel-gomes-tenda-digital-66868648)
_
====
[](http://www.slideshare.net/NodejsFoundation/from-pterodactyls-and-cactus-to-artificial-intelligence-ivan-seidel-gomes-tenda-digital-66868648)
_
====
### Genetic Algorithm
[](http://www.slideshare.net/NodejsFoundation/from-pterodactyls-and-cactus-to-artificial-intelligence-ivan-seidel-gomes-tenda-digital-66868648)
_
====
[](http://www.slideshare.net/NodejsFoundation/from-pterodactyls-and-cactus-to-artificial-intelligence-ivan-seidel-gomes-tenda-digital-66868648)
_
====
[](http://www.slideshare.net/NodejsFoundation/from-pterodactyls-and-cactus-to-artificial-intelligence-ivan-seidel-gomes-tenda-digital-66868648)
_
====
[](http://www.slideshare.net/NodejsFoundation/from-pterodactyls-and-cactus-to-artificial-intelligence-ivan-seidel-gomes-tenda-digital-66868648)
_
====
[](http://www.slideshare.net/NodejsFoundation/from-pterodactyls-and-cactus-to-artificial-intelligence-ivan-seidel-gomes-tenda-digital-66868648)
_
====
[](http://www.slideshare.net/NodejsFoundation/from-pterodactyls-and-cactus-to-artificial-intelligence-ivan-seidel-gomes-tenda-digital-66868648)
_
====
[](http://www.slideshare.net/NodejsFoundation/from-pterodactyls-and-cactus-to-artificial-intelligence-ivan-seidel-gomes-tenda-digital-66868648)
----
## Demo T-Rex
<iframe frameborder=0 width="630" height="170" marginheight=0 marginwidth=0 scrolling="auto" src="http://www.pepitosoft.com/demos/trex/index.html"></iframe>
[T-Rex](https://github.com/amaneureka/T-Rex/)
_
====
### Chrome/Chromium offline script
[](http://caza.la/synaptic/#/)
- [Chromium](https://chromium.googlesource.com/chromium/src.git/+/master/components/neterror/resources/offline.js)
- [Code Google.](https://code.google.com/p/chromium/codesearch#chromium/src/components/neterror/resources/offline.js&q=offline.js&sq=package:chromium&type=cs)
_
====
### Synaptic
[](http://caza.la/synaptic/#/)
----
## Useful, Reference and Future links
- [BlueMix](https://www.google.com.co/aclk?sa=L&ai=DChcSEwiDvJ715rDQAhVXG4EKHXicAzgYABAA&sig=AOD64_2dC71QClj6e8QGg02JsbBbGDHEsQ&q=&ved=0ahUKEwi695j15rDQAhUF5SYKHbP8BP4Q0QwIFw&adurl=)
- [Chromium](https://chromium.googlesource.com/chromium/src.git/+/master/components/neterror/resources/offline.js)
- [opencv](http://docs.opencv.org/2.4/doc/tutorials/introduction/display_image/display_image.html)
- [robotjs](https://github.com/octalmage/robotjs)
- [IAMDinosaur](https://github.com/ivanseidel/IAMDinosaur)
- [synaptic](http://caza.la/synaptic/#/)
[](http://oems-vis-rec-demo.mybluemix.net/)
_
====
## Useful, Reference and Future links
- [Code Google.](https://code.google.com/p/chromium/codesearch#chromium/src/components/neterror/resources/offline.js&q=offline.js&sq=package:chromium&type=cs)
- [wikitude](http://www.wikitude.com/blog-migrating-moodstocks-heres-keep-app-online/)
- [moodstocks](https://moodstocks.com/)
- [tngan](https://github.com/tngan)
- [T-Rex](https://github.com/amaneureka/T-Rex/)
[](http://oems-vis-rec-demo.mybluemix.net/)
_
====
## Useful, Reference and Future links
- [convnetjs](http://cs.stanford.edu/people/karpathy/convnetjs/)
- [THE MNIST DATABASE](http://yann.lecun.com/exdb/mnist/)
- [Neural Networks in Javascript](http://blog.webkid.io/neural-networks-in-javascript/)
- [Top Machine Learning Libraries for Javascript](http://www.kdnuggets.com/2016/06/top-machine-learning-libraries-javascript.html)
[](http://oems-vis-rec-demo.mybluemix.net/)
_
====
## Useful, Reference and Future links
- [Playing Atari with Deep Reinforcement Learning](https://arxiv.org/pdf/1312.5602v1.pdf)
- [Understanding LSTM Networks](http://colah.github.io/posts/2015-08-Understanding-LSTMs/)
- [5 Machine Learning Projects You Can No Longer Overlook](http://www.kdnuggets.com/2016/05/five-machine-learning-projects-cant-overlook.html)
- [Top 10 Machine Learning Projects on Github](http://www.kdnuggets.com/2015/12/top-10-machine-learning-github.html/2)
[](http://oems-vis-rec-demo.mybluemix.net/)
_
====
## Useful, Reference and Future links
- [Lian Li: Machine Learning with Node.js - JSUnconf 2016](https://www.youtube.com/watch?v=M5glN6XjDv8)
- [Machine learning is not the future - Google I/O 2016](https://www.youtube.com/watch?v=3dXQxSI3XDY)
- [Machine Learning Tutorial for JavaScript Hackers: Tips On Gathering and Pre-Processing Data](https://www.youtube.com/watch?v=5AGM7grR680)
- [ConvNetJS – Deep Learning in your browser](https://www.youtube.com/watch?v=nAHcrz5hxc4)
[](http://oems-vis-rec-demo.mybluemix.net/)
_
====
## Useful, Reference and Future links
- [Deep Learning Frameworks Compared](https://www.youtube.com/watch?v=MDP9FfsNx60)
- [7.1 Recurrent Neural Networks RNN](https://www.youtube.com/watch?v=kMLl-TKaEnc)
- []()
- []()
[](http://oems-vis-rec-demo.mybluemix.net/)
_
====
## Useful, Reference and Future links
- [IBM Watson: Using the Visual Recognition service ](https://www.youtube.com/watch?v=FYZld6SSCnY)
- [Neural Networks in JavaScript](http://www.antoniodeluca.info/blog/10-08-2016/neural-networks-in-javascript.html)
- [Introduction to Artificial Neural Networks](http://www.theprojectspot.com/tutorial-post/introduction-to-artificial-neural-networks-part-1/7)
- [A 'Brief' History of Neural Nets and Deep Learning](http://www.andreykurenkov.com/writing/a-brief-history-of-neural-nets-and-deep-learning/)
[](http://oems-vis-rec-demo.mybluemix.net/)
_
====
## Useful, Reference and Future links
- [Synaptic - Advanced Neural Nets In JavaScript](http://www.i-programmer.info/news/105-artificial-intelligence/7890-synaptic-advanced-neural-nets-in-javascript.html)
- [A Beginner’s Guide to Recurrent Networks and LSTMs](https://deeplearning4j.org/lstm)
- []()
- []()
[](http://oems-vis-rec-demo.mybluemix.net/)
_
====
## Useful, Reference and Future links
- []()
- []()
- []()
- []()
[](http://oems-vis-rec-demo.mybluemix.net/)