Introduction to

Data Science Deep Learning & AI

Introduction to Deep Learning & AI &
Machine Learning Big Data
 What is Big Data?
Deep Learning: A revolution in Artificial  5 Vs of Big Data
Intelligence  Big Data Architecture
 Limitations of Machine Learning  Big Data Technologies
What is Deep Learning?  Big Data Challenge
 Need for Data Scientists  Big Data Requirements
 Foundation of Data Science  Big Data Distributed Computing &
 What is Business Intelligence Complexity
 What is Data Analysis  Hadoop
 What is Data Mining  Map Reduce Framework
What is Machine Learning  Hadoop Ecosystem
Analytics vs Data Science
 Value Chain Data Science Deep Dive
 Types of Analytics  What Data Science is
 Lifecycle Probability  Why Data Scientists are in demand
 Analytics Project Lifecycle  What is a Data Product
 Advantage of Deep Learning over  The growing need for Data Science
Machine learning  Large Scale Analysis Cost vs Storage
 Reasons for Deep Learning  Data Science Skills
 Real-Life use cases of Deep Learning  Data Science Use Cases
 Review of Machine Learning  Data Science Project Life Cycle &
Data Stages
 Basis of Data Categorization  Data Acuqisition
 Types of Data  Where to source data
 Data Collection Types  Techniques
 Forms of Data & Sources  Evaluating input data
 Data Quality & Changes  Data formats
 Data Quality Issues  Data Quantity
 Data Quality Story  Data Quality
 What is Data Architecture  Resolution Techniques
 Components of Data Architecture  Data Transformation
 OLTP vs OLAP  File format Conversions
 How is Data Stored?  Annonymization
 Statistics
What is Statistics

Descriptive Statistics
 Central Tendency Measures
 The Story of Average
 Dispersion Measures
 Data Distributions

Central Limit Theorem

 What is Sampling

Why Sampling
 Sampling Methods

Inferential Statistics

What is Hypothesis testing

 Confidence Level
 Degrees of freedom
 what is pValue

Chi-Square test

What is ANOVA
 Correlation vs Regression

Uses of Correlation & Regression

Getting Started with Python  Generator Expressions
 Python Overview  Dictionaries and Sets.
 About Interpreted Languages Numpy & Pandas with Matplotlib &
 Advantages/Disadvantages of Seaborn
Python pydoc.  Learning NumPy
 Starting Python  Pllotting using Matplotlib and
 Interpreter PATH Seabron
 Using the Interpreter  Machine Learning application
 Running a Python Script  Introduction to Pandas
 Using Variables  Creating Data Frames
 Keywords  GroupingSorting
 Built-in Functions  Plotting Data
 StringsDifferent Literals  Creating Functions
 Math Operators and Expressions  Converting Different Formats
 Writing to the Screen  Combining Data from Various
 String Formatting Formats
 Command Line Parameters and Flow  Slicing/Dicing Operations.
Control.
Deep Dive - Functions Sorting Errors and
Sequences and File Operations Exception Handling
 Lists  Functions
 Tuples  Function Parameters
 Indexing and Slicing  Global Variables
 Iterating through a Sequence  Variable Scope and Returning
 Functions for all Sequences Values. Sorting
 Using Enumerate()  Alternate Keys
 Operators and Keywords for  Lambda Functions
Sequences  Sorting Collections of Collections
 The xrange() function  Sorting Dictionaries
 List Comprehensions  Sorting Lists in Place
 Errors and Exception Handling
  Handling Multiple Exceptions  Introduction to Python Classes
 The Standard Exception Hierarchy  Defining Classes
 Using Modules  Initializers
 The Import Statement  Instance Methods
 Module Search Path  Properties
 Package Installation Ways.  Class Methods and DataStatic
Methods
Regular Expressionsit's Packages and  Private Methods and Inheritance
Object  Module Aliases and Regular
 Oriented Programming in Python Expressions.
 The Sys Module
 Interpreter Information Debugging, Databases and Project
 STDIO Skeletons
 Launching External Programs  Debugging
 PathsDirectories and Filenames  Dealing with Errors
 Walking Directory Trees  Creating a Database with SQLite 3
 Math Function  CRUD Operations
 Random Numbers  Creating a Database Object.
 Dates and Times
 Zipped Archives
 Plotly & Dash
Getting Started with Plotly  Controlling Callbacks with Dash
 Plotly and Dash Overview State
 Plotly Basics  Interacting with Visualizations
 Scatter Plots  Hover Over Data
 Line Charts  Click Data
 Bar Charts  Selection Data
 Bubble Plots  Updating Graphs on Interactions
 Box Plots  Updating Graphs on Interactions
 Histograms Part 2
 Distplots  Updating Graphs on Interactions -
 Heatmaps Part Three
 Project Imports and Graph Setup
Introduction to Dash  Input Box and Basic Callback
 Dash Basics - Layout  Reading Data with Pandas
 Introduction to Dash Basics Datareader
 Dash Layouts  Adding DatePickers for Choosing
 Dash Layouts - Styling Dates
 Converting Simple Plotly Plot to  Adding in Dash State
Dashboard with Dash  Multiple Stock Option Dropdown
 DashBoard Basics  Live Updating
 Create a Simple Dashboard  Layout Updating
 DashBoard Components  Deployment
 Dash Components  App Authorization
 HTML Components  Deploying App to Heroku
 Core Components
 Markdown with Dash
 Interactive Components
 Single Callbacks for Interactivity
 Dash Callbacks for Graphs
 Multiple Inputs
 Multiple Outputs
 Callbacks with State
 Machine Learning Deep Learning & AI
using Python
Introduction  Understanding K-Means Clustering
 ML Fundamentals  Understanding TF-IDF, Cosine
 ML Common Use Cases Similarity and their application to
 Understanding Supervised and Vector Space Model
Unsupervised Learning Techniques  Case study

Linear Regression Implementing Association rule mining

Case study in Retail Domain
Case study in Financial Domain  What is Association Rules & its use
 Introduction to Predictive Modeling cases?
in Financial Domain  What is Recommendation Engine &
 Linear Regression Overview it’s working?
 Simple Linear Regression  Recommendation Use-case
 Multiple Linear Regression  Case study

Logistic Regression Understanding Process flow of Supervised

Learning Techniques
Case study in Banking Domain
 Logistic Regression Overview Decision Tree Classifier
 Data Partitioning
 Univariate Analysis Case study in Healthcare Domain
 Bivariate Analysis  How to build Decision trees
 Multicollinearity Analysis  What is Classification and its use
 Model Building cases?
 Model Validation  What is Decision Tree?
 Model Performance Assessment  Algorithm for Decision Tree
AUC & ROC curves Induction
 Scorecard  Creating a Decision Tree
 Confusion Matrix
Clustering  Case study
Case study in Ecommerce Domain
 Similarity Metrics Random Forest Classifier
 Distance Measure Types: Euclidean,  What is Random Forests
Cosine Measures  Features of Random Forest
 Creating predictive models
  Out of Box Error Estimate and  Forecast the data using the
Variable Importance respective model
 Case study in Healthcare Domain  What is Time Series data?
 Time Series variables
Naive Bayes Classifier.  Different components of Time Series
 Case study data
 Visualize the data to identify Time
Project Discussion Series Components
Problem Statement and Analysis  Implement ARIMA model for
 Various approaches to solve a Data forecasting
Science Problem  Exponential smoothing models
 Pros and Cons of different  Identifying different time series
approaches and algorithms. scenario based on which different
Exponential Smoothing model can
Support Vector Machines be applied
Case study in Healthcare Domain  Implement respective model for
 Introduction to SVMs forecasting
 SVM History  Visualizing and formatting Time
 Vectors Overview Series data
 Decision Surfaces  Plotting decomposed Time Series
 Linear SVMs data plot
 The Kernel Trick  Applying ARIMA and ETS model for
 Non-Linear SVMs Time Series forecasting
 The Kernel SVM  Forecasting for given Time period
 Case Study
Time Series Analysis
Case study in Stock Exchange & Power Project Machine learning algorithms
Infra Domain Python
 Describe Time Series data  Various machine learning algorithms
 Format your Time Series data in Python
 List the different components of  Apply machine learning algorithms
Time Series data in Python
 Discuss different kind of Time Series
scenarios Feature Selection and Preprocessing
 Choose the model according to the  How to select the right data
Time series scenario  Which are the best features to use
 Implement the model for  Additional feature selection
forecasting techniques
 Explain working and implementation  A feature selection case study
of ARIMA model  Preprocessing
 Illustrate the working and  Preprocessing Scaling Techniques
implementation of different ETS  How to preprocess your data
models  How to scale your data
  Feature Scaling Final Project  A final ensemble technique

Which Algorithms perform best Model selection cross validation score

 Highly efficient machine learning  Introduction Model Tuning
algorithms  Parameter Tuning GridSearchCV
 Bagging Decision Trees  A second method to tune your
 The power of ensembles algorithm
 Random Forest Ensemble technique  How to automate machine learning
 Boosting - Adaboost  Which ML algo should you choose
 Boosting ensemble stochastic  How to compare machine learning
gradient boosting algorithms in practice
 Deep Learning & AI using Python
Deep Learning & AI  Different variants of Gradient
 "Case study in Stock Exchange & Descent
 Image Recognition & Banking  Case study inBanking Domain
Domain"
 Deep Learning Overview Convolutional Neural Networks
 The Brain vs Neuron  Case study inImage Recognition
 Introduction to Deep Learning Domain
 Convolutional Operation
Introduction to Artificial Neural Networks  Relu Layers
 The Detailed ANN  What is Pooling vs Flattening
 The Activation Functions  Full Connection
 How do ANNs work & learn  Softmax vs Cross Entropy
 Gradient Descent  " Building a real world convolutional
 Stochastic Gradient Descent neural network
 Backpropogation  for image classification"
 Understand limitations of a Single
Perceptron What are RNNs - Introduction to RNNs
 Understand Neural Networks in  Case study inStock Exchange
Detail Domain
 Illustrate Multi-Layer Perceptron  Recurrent neural networks rnn
 Backpropagation – Learning  LSTMs understanding LSTMs
Algorithm  long short term memory neural
 Understand Backpropagation – networks lstm in python
Using Neural Network Example
 MLP Digit-Classifier using Restricted Boltzmann Machine (RBM) and
TensorFlow Autoencoders
 Building a multi-layered perceptron  Restricted Boltzmann Machine
for classification  Applications of RBM
 Why Deep Networks  Introduction to Autoencoders
 Why Deep Networks give better  Autoencoders applications
accuracy?  Understanding Autoencoders
 Use-Case Implementation  Building a Autoencoder model
 Understand How Deep Network
Tensorflow with Python
Works?
 Introducing Tensorflow
 How Backpropagation Works?
 Introducing Tensorflow
 Illustrate Forward pass, Backward
 Why Tensorflow?
pass
 What is tensorflow?
  Tensorflow as an Interface  ConvNet Architecture
 Tensorflow as an environment  Overfitting and Regularization
 Tensors  Max Pooling and ReLU activations
 Computation Graph  Dropout
 Installing Tensorflow  Strides and Zero Padding
 Tensorflow training  Coding Deep ConvNets demo
 Prepare Data  Debugging Neural Networks
 Tensor types  Visualising NN using Tensorflow
 Loss and Optimization  Tensorboard
 Running tensorflow programs
Transfer Learning using Keras and TFLearn
Building Neural Networks using Tensorflow  Transfer Learning Introduction
 Tensors  Google Inception Model
 Tensorflow data types  Retraining Google Inception with our
 CPU vs GPU vs TPU own data demo
 Predicting new images
 Tensorflow methods
 Transfer Learning Summary
 Introduction to Neural Networks
 Extending Tensorflow
 Neural Network Architecture  Keras
 Linear Regression example revisited  TFLearn
 The Neuron  Keras vs TFLearn Comparison
 Neural Network Layers
 The MNIST Dataset Text Mining& NLP & Deep NLP & Chatbot
 Coding MNIST NN  Case study
Sentimental Analysis
Deep Learning using Tensorflow  Case study
 Deepening the network Computer Vision
 Images and Pixels  Case study
 How humans recognise images
 Convolutional Neural Networks
 Computer Vision using Open CV
NumPy and Image Basics Corner Detection - Harris Corner Detection
Introduction to Numpy and Image Section  Corner Detection - Shi-Tomasi
 NumPy Arrays Detection
 What is an image?
 Images and NumPy Edge Detection
 Image Basics with OpenCV Grid Detection
 Contour Detection
Introduction to Images and OpenCV Basics Feature Matching
 Opening Image files with OpenCV  Feature Matching
 Drawing on Images Watershed Algorithm
 Custom Seeds with Watershed
Direct Drawing on Images with a mouse - Algorithm
Advanced Introduction to Face Detection
 Image Processing Face Detection with OpenCV
 Color Mappings  Detection Assessment
 Blending and Pasting Images Object Tracking
Introduction to Object Tracking
Blending and Pasting Images - Masks -  Optical Flow
Advanced Optical Flow Coding with OpenCV
 Image Thresholding MeanShift and CamShift Tracking Theory
 Blurring and Smoothing  MeanShift and CamShift Tracking
with OpenCV
Blurring and Smoothing - Advanced  Overview of various Tracking API
Morphological Operators Methods
Gradients  Tracking APIs with OpenCV
 Histograms Deep Learning for Computer Vision
Histogram Eqaulization Understanding Classification Metrics
 Image Processing Assessment  Introduction to YOLO v3
 YOLO Weights Download
Video Basics with Python and OpenCV  YOLO v3 with Python
 Introduction to Video Basics
 Connecting to Camera
Using Video Files
Drawing on Live Camera
 Video Basics Assessment
Object Detection with OpenCV and Python
 Introduction to Object Detection
 Template Matching
 NLP - Chatbots Text Mining
Natural Language Processing Basics
Introduction to Natural Language Processing
 What is Natural Language Processing?
 Tokenization - Part One
 Stemming
 Lemmatization
 Stop Words
 Phrase Matching and Vocabulary
 Part of Speech Tagging and Named Entity Recognition

Introduction to Section on POS and NER

 Part of Speech Tagging

Named Entity Recognition

Sentence Segmentation
 Text Classification
Introduction to Text Classification
 Classification Metrics
Confusion Matrix
Text Feature Extraction
Semantics and Sentiment Analysis
 Overview of Semantics and Word Vectors
 Semantics and Word Vectors with Spacy

Sentiment Analysis with NLTK

 Topic Modeling
 Latent Dirichlet Allocation Overview

Deep Learning for NLP

 The Basic Perceptron Model
 Introduction to Neural Networks
Keras Basics
Recurrent Neural Network
LSTMs, GRU, and Text Generation

Chat Bots Overview

 Creating Chat Bots with Python
 Intro to R Programming
Introduction to R  Packages installation for database
 Business Analytics import
 Analytics concepts  Connecting to RDBMS from R using
 The importance of R in analytics ODBC and basic SQL queries in R
 R Language community and eco-  Web Scraping
system  Other concepts on Data Import
 Usage of R in industry Techniques
 Installing R and other packages
 Perform basic R operations using Exploratory Data Analysis (EDA) using R
command line  What is EDA?
 Usage of IDE R Studio and various  Why do we need EDA?
GUI  Goals of EDA
 Types of EDA
R Programming Concepts  Implementing of EDA
 The datatypes in R and its uses  Boxplots, cor() in R
 Built-in functions in R  EDA functions
 Subsetting methods  Multiple packages in R for data
 Summarize data using functions analysis
 Use of functions like head(), tail(),  Some fancy plots
for inspecting data  Use-cases for EDA using R
 Use-cases for problem solving using
R Data Visualization in R
 Story telling with Data
Data Manipulation in R  Principle tenets
 Various phases of Data Cleaning  Elements of Data Visualization
 Functions used in Inspection  Infographics vs Data Visualization
 Data Cleaning Techniques  Data Visualization & Graphical
 Uses of functions involved functions in R
 Use-cases for Data Cleaning using R  Plotting Graphs
 Customizing Graphical Parameters
Data Import Techniques in R to improvise the plots
 Import data from spreadsheets and  Various GUIs
text files into R  Spatial Analysis
 Importing data from statistical  Other Visualization concepts
formats
 Apache Spark using Scala
Apache Spark  Spark Execution Context
 Introduction to Apache Spark  What is RDD
 Why Spark  RDD Deep Dive
 Batch Vs. Real Time Big Data  RDD Dependencies
Analytics  RDD Lineage
 Batch Analytics - Hadoop Ecosystem  Spark Application In Depth
Overview,  Spark Deployment
 Real Time Analytics Options,  Parallelism in Spark
 Streaming Data - Storm,  Caching in Spark
 In Memory Data - Spark, What is
Spark Internals & Spark SQL
Spark?,
 Spark Transformations
 Spark benefits to Professionals
 Spark Actions
 Limitations of MR in Hadoop
 Spark Cluster
 Components of Spark
 Spark SQL Introduction
 Spark Execution Architecture
 Spark Data Frames
 Benefits of Apache Spark
 Spark SQL with CSV
 Hadoop vs Spark
 Spark SQL with JSON
Spark Core Architecture  Spark SQL with Database
 Spark & Distributed Systems Introducing Mllib
 Spark for Scalable Systems
 HADOOP
Big Data and Hadoop Introduction  HiveQL: Data Types, Operators and
 What is Big Data and Hadoop? Functions,
 Challenges of Big Data  Hands-On Exercise
 Traditional approach Vs Hadoop  Hive Tables ,Managed Tables and
 Hadoop Architecture External Tables
 Distributed Model  Hands-On Exercise
 Block structure File System  Partitions and Buckets
 Technologies supporting Big Data  Hands-On Exercise
 Replication  Storage Formats, Importing Data,
 Fault Tolerance Altering Tables, Dropping Tables
 Why Hadoop?  Hands-On Exercise
 Hadoop Eco-System  Querying Data, Sorting and
 Use cases of Hadoop Aggregating, Map Reduce Scripts,
 Fundamental Design Principles of  Hands-On Exercise
Hadoop  Joins & Sub queries, Views
 Comparison of Hadoop Vs RDBMS  Hands-On Exercise
 When to Use HIVE, Impala and Pig
Map Reduce Concepts  Hands on Exercises
 What is Map Reduce?  Integration, Data manipulation with
 Why Map Reduce? Hive
 Map Reduce in real world.  Hands-On Exercise
 Map Reduce Flow  User Defined Functions,
 What is Mapper?  Hands-On Exercise
 What is Reducer?  Appending Data into existing Hive
 What is Shuffling? Table
 Word Count Problem  Hands-On Exercise
 Distributed Word Count Flow &  Static partitioning vs dynamic
Solution partitioning
 Log Processing and Map Reduce  Hands-On Exercise

HIVE
 Hive Fundamentals & Architecture
 Loading and Querying Data in Hive
 Hands-On Exercise
 Hive Architecture and Installation
 Comparison with Traditional
Database
Azure Machine Learning  Logistic Regression - Understand
 Azure Machine Learning Workflow Parameters and Their Impact
 Getting Access to Azure Machine Learning  Understanding the Confusion Matrix, AUC,
 Azure Machine Learning Studio Accuracy, Precision, Recall and F1Score
 Creating Models using Azure ML  Logistic Regression - Understanding the
 Getting and Exploring Data using Azure ML results
 Selecting the Algorithm and Training the  Logistic Regression - Model Selection and
Model Impact Analysis
 Evaluating the Trained Model  Logistic Regression - Build Multi-Class Wine
 Deep into Azure Machine Learning Quality Prediction Model
 Decision Tree - What is Decision Tree?
Data Processing  Decision Tree - Ensemble Learning -
 Data Input-Output - Upload Data Bagging and Boosting
 Data Input-Output - Convert and Unpack  Two Class Decision Forest - Income
 Data Input-Output - Import Data Prediction
 Data Transform - Add Rows/Columns,  Decision Tree - Multi Class Decision Forest
Remove Duplicates, Select Columns  SVM - What is Support Vector Machine?
 Data Transform - Apply SQL  SVM - Census Income Prediction
Transformation, Clean Missing Data, Edit  Hyperparameter Tuning
Metadata  Tune Hyperparameter for Best Parameter
 Sample and Split Data - How to Partition or Selection
Sample, Train and Test Data
Regression Analysis
 Classification  What is Linear Regression?
 Logistic Regression - What is Logistic  Regression Analysis - Common Metrics
Regression?  Linear Regression model using OLS
 Logistic Regression - Build Two-Class Loan  Linear Regression - R Squared
Approval Prediction Model  Gradient Descent
Clustering Data to Azure Machine Learning
 What is Cluster Analysis?  Getting and Exploring Data using Azure ML
 Cluster Analysis Experiment 1  Data Pre-processing using Azure ML
 Cluster Analysis Experiment 2 - Score and  Incorporating Code
Evaluate  Selecting and Training an Algorithm
 Evaluating Your Trained Model
Recommendation System  Understanding Evaluation
 What is a Recommendation System?  Comparing Model Results
 Data Preparation using Recommender Split
 What is Matchbox Recommender and Deploy WebService
Train Matchbox Recommender  Azure ML Webservice - Prepare the
 How to Score the Matchbox experiment for webservice
Recommender?  Deploy Machine Learning Model As a Web
 Restaurant Recommendation Service
 Understanding the Recommendation  Use the Web Service
Results