Introduction of Deep Learning

Lecture: 02
 Perceptron algorithms

Perceptron Learning Steps

Introduction to Machine Learning Software

Types of Machine Learning Software

Merits of Machine Learning Software

Applications Machine Learning

Agenda
Introduction to Deep Learning Toolbox

Deep Learning Toolbox in Detail

Types of Convolutional Neural Networks

Classification Task in MATLAB

Benefits of Deep Learning Toolbox

  Perceptron algorithms can be divided into two types they are
single layer perceptrons and multi-layer perceptron’s. In single-layer
 Perceptron perceptron’s neurons are organized in one layer whereas in a multilayer
perceptron’s a group of neurons will be organized in multiple layers.
algorithms Every single neuron present in the first layer will take the input signal
and send a response to the neurons in the second layer and so on
 Single Layer
Perceptron
Multi-Layer
Perceptron
  Features of the model we want to train should be passed
as input to the perceptrons in the first layer.
 These inputs will be multiplied by the weights or weight
coefficients and the production values from all
perceptrons will be added.
 Adds the Bias value, to move the output function away
Perceptron from the origin.
Learning Steps  This computed value will be fed to the activation function
(chosen based on the requirement, if a simple
perceptron system activation function is step function).
 The result value from the activation function is the
output value.
  Machine Learning is a scientific application based on AI(Artificial
Intelligence) that enables the computer system to learn, perform a
certain task, and improve the experience gained out of it without
Introduction to programming it actually exclusively. Machine Learning algorithms are
Machine Learning used in day-to-day exercises such as search engines, image & speech
recognition, healthcare diagnosis, etc. These algorithms need certain
Software software and tools to execute and perform certain actions. Machine
Learning software has its own set of tools, libraries, and resources to
develop applications that can replicate human behavior with smart
machines and applications.
  There are tons of Machine Learning software available in the market.
Some of the software has exclusive and unique features. At the same
time, there are some tools that are comparatively better. Many
businesses have started to adopt the technology to increase their ROI
(Return on Investment).
 As Machine Learning is a part of AI (Artificial Intelligence), machines
Types of Machine are trained from the result or pattern to develop new changes
Learning Software autonomously.
 Machine Learning software are- Azure Machine Learning Studio,
Shogun, Apache Mahout, Apache Spark MLlib, IBM Watson Machine
Learning, RapidMinor, Weka, Google Cloud ML Engine, Pytroch,
Figure Eight, Crab, Microsoft Cognitive Toolkit, Torch, etc. Let’s have a
closer look at some top and most used software that Machine Learning
experts rely on.
  Google Cloud ML Engine
 Google’s Cloud Machine Learning Engine is one of the popular
applications for training, analysis, and deep learning.
 Azure ML Studio
 It is a code-free, drag and drop solution for ML experts for convenient
use made by Microsoft.
Types of Machine
 IBM Watson Machine Learning
Learning Software
 With its open-source model operation, it helps data scientists and
developers to expedite AI and Machine Learning applications.
 TensorFlow
 It’s a new open-source framework released in 2015 and not complex to
operate and as well can be deployed over various platforms. Made by
Google, TensorFlow is available for Python, C++, Java, Rust, etc.
  Microsoft Cognitive Toolkit
 Microsoft’s newly launched AI solution can train the machine with its
deep learning algorithms to behave like a human brain and face. It can handle data
from Python, C++, etc.
 Theano
 It is a Python-based open-source library for deep learning algorithms to accelerate
Machine Learning deployment. Theano is capable of taking data structures to the
Types of Machine next level with the integration of Python libraries like NumPy, Pandas, and some
Learning Software native programs.
Torch
 This typical older Machine Learning library fascinates flexibility and speediness
in operations while deploying Machine Learning projects.
 Apache Spark MLlib
 Apache Spark is a trusted Machine Learning tool for advanced Machine Learning
and deep learning. It is a seamless, scalable platform that can be integrated with
Hadoop for better results from algorithms. Classification, Regression, Gradient
boosting, Decision Trees, LDA, etc are some of the algorithms to support MLlib.
  Pytorch
 Pytorch is developed by Facebook for advanced deep learning using
Neural Networks and Tensors. It is ascertained that researchers across
the world use Pytorch for dynamic graphical representation and
Types of Machine blueprint techniques.
 Ideas2T Technologies
Learning Software
 Ideas2T tool, unlike others, has a unique feature that can help recruiters
to hire the right candidates based on their resume(s) through Machine
Learning. It has been an intersection for start-ups, enterprises and
business houses because of its cutting-edge and advanced algorithms.
  Machine Learning, a subset of AI that helps to examine and learn from
the data and to make certain decisions out of it. It is capable of
delivering decisions and recommendations at high speed with accuracy.
So, ML software is very popular. Interestingly, they are inexpensive
too. Both Machine Learning and AI work on large and complex data
sets to visualize and make perfect decisions.
Merits of Machine  A few examples are as under:
Learning Software • Facebook: When we upload a photo on Facebook, it recognizes a
person from the same photo and suggests mutual friendship. This is
how ML works.
• Netflix: Netflix sometimes sends suggestions of web shows or movies
based on what we have watched earlier. Basically, Machine Learning is
used to select data based on choice.
 • Financial Services: The financial sector use this to identify insights for
investments, trading and even ascertaining financial risks.
• Marketing & Sales: Various companies and other similar
establishments use Machine Learning software to analyze the history of
Applications purchases and make recommendations for customers depending on it.
Machine Learning • Healthcare: Electronic wearables and sensors are used widely
nowadays. Sensors in those wearables provide real-time data on blood
pressure, heartbeat, and other vital information related to health.
 • Transportation: Machine Learning software analyses the travel history
and routes to identify problems, if any, so as to help
customers/passengers opt-out of risk-prone areas in the future.
• Government: Some Government agencies use ML tools to extract
insights to minimize cost and increase efficiency in their operations.
Applications
• Oil & Gas: Machine Learning is mostly used in the energy sectors. It
Machine Learning helps analyze existing and new energy sources for exploration and
distribution. Many oil agencies in both Government and Private sectors
have adopted these techniques.
  Deep Learning Toolbox, a framework developed by the MathWorks is
used in the development of deep neural networks. It supports advanced
architectures like Convolution Neural Networks, Generative Adversarial
Network, Siamese Networks, etc. which finds its application in image,
Deep Learning video and text processing. It also provides an interactive dashboard for
Toolbox plotting and visualizing the network architecture. It also supports model
exchange between TensorFlow and PyTorch by using the ONNX
format. It also supports parallel GPUs through the usage of Parallel
Computing Toolbox which uses a scalable architecture for supporting
the cloud and cluster platform which includes Amazon EC2 instance,
NVIDIA, etc.
Deep Learning
Toolbox in
Detail
  Deep Learning Toolbox uses MATLAB as its backend programming
language. Thus, the users can take reference of the MATLAB Deep
Neural Network. It also provides a MATLAB Coder also known as
GPU coder which can be used in the generation of the C++ and CUDA
code which can be deployed on Intel, NVIDIA and ARM platforms. It
also comes up with a Deep Network Designer which eases the
Deep Learning development effort on part of the user. The Deep Network Designer
comes up with an easy drag-drop method that helps the user in
Toolbox in Detail designing, developing and visualizing his/her application easily. It also
provides the user with a method to apply pre-trained models as well as
to develop new model architectures. Also, MATLAB Image Labeler
application can be used along with it so that the user can get the benefit
of viewing his/her images as well as labeling them for semantically
segmenting them. Using the above, users can create a workflow where
they can establish ground truth for various domain-specific
applications. It also supports the fine-tuning of the pre-trained models.
  The Deep Learning Toolbox supports the following types of
Convolutional Neural Networks:
 MatConvNet
 It is one of the processes used by the MATLAB for the convolutional
Types of neural networks. It is simple and flexible in terms of usage. It
comprises all the necessary setup required for building the CNN
Convolutional architecture with inbuilt MATLAB functions, calculation of
Neural Networks convolutions, pooling, etc. It also supports the quick and dirty
implementation of the distinct neural network architectures that can be
developed for faster testing. It also supports computation which use
CPU as well as GPU for training big and complex models for any large
dataset.
  Cuda-Convnet
 It is an implementation that is secured and uses C++/CUDA as its
backend. It presents the user with the connectivity in the optional layers
Types of as well as the intensity of networks. It accepts any acyclic graph of
Convolutional layers as its input if marked. The model is trained using back-
Neural Networks propagation. The method of back-propagation involves the gradient
computation for a neural network which calculates the underlying
weights of the network. The error is calculated at the output layer which
is propagated back to the other layers. It is used by CUDA as it is a
generalized form of the delta approach which finds its usage in multi-
layered feedforward networks.
  Below is an implementation of the AlexNet image classification task in
MATLAB. Only 8 lines of code are required to define the above task.
 Code
Classification Task
in MATLAB
 • MATLAB: It comes up with MATLAB as its back-end which finds its
application in IoT (Internet of Things) applications, simulation
software, etc.
• Applications: It supports Machine Learning and Deep Learning
technology in the existing framework. It supports the regression as well
as classification tasks including forecasting, time series analysis, image
Benefits of Deep processing, text processing, audio processing etc.
Learning Toolbox • ONNX: It uses ONNX (Open Neural Network Exchange Format)
which allows the users to export and import their existing models which
are developed and deployed using existing frameworks like
TensorFlow, Keras, Caffe, MXNet etc.
• DNN Support: Deployment code for DNN can be used on multiple
platforms like NVIDIA Tegra, Intel, ARM, etc.
 • Deep Network Designer: Pre-Trained models like AlexNet, Google
LeNet, ResNet etc. can be used by users which can be fine-tuned.
Workflows can be created based on domain which help in establishing
the ground truth labels for the applications that use image, video and
speech segments.
Benefits of Deep • Distributed Computation: It provides Parallel Computing Toolbox
Learning Toolbox which uses the distributed computing architecture for training complex
models for a huge dataset. The MATLAB distributed Computing
Server supports the GPU instances provided by the Amazon EC2
cloud. It also supports other popular cloud-based platforms like Azure
and Amazon AWS.
• Other Features: 5G Toolbox, NVIDIA based cloud platform, DGX
based Sensor Fusion/Tracking, etc.
  https://www.educba.com/deep-learning/
 https://www.educba.com/deep-learning-toolbox/
 https://www.educba.com/deep-learning-software
Supporting Material  https://towardsdatascience.com
 https://www.edureka.com
 https://www.geeksforgeeks.org
Q/A Thanks