GREAT LEARNING COURSES CURRICULUM VITAE

WEEK – 1
ITP
DAY-1
 Intro to Python –
 Basic Commands - Hello World
 Variables
 Basic Arithmetic & logical operators (int, float)
 Data Types - int, float, strings
 Concat, Subset, Position, length etc.
 Appreciation of programming using Pseudo Code (Introduction) - If-else, loops (Deck)

Lab Exercises (2 hrs)

DAY-2
 List, Tuples, Dictionaries
 Indexing
 Arithmetic operators
 Logical operators
 Comparison operators

 Intro to Conditional statements (if-else, elif), Nested Conditional

 Intro to Basic For, While Loops, Break

Lab Exercises (2 hrs)

DAY-3
 psuedo codes into programs using Loops and if-else
 List Comprehension
 Use cases
 vs Loops

Lab Exercises (2 hrs)

DAY-4
 Understanding the concept of functions
 Exploring commonly used built in functions (min, max, sort etc.)
 Programming user defined functions
 Working with functions with and without arguments
 Functions with return items
 Understanding lambda functions
 Overview of map, reduce and filter functions

Lab Exercises - Mini Case Study

 WEEK – 2
Numpy,Pandas,Visualization
DAY-1
 Indexing and slicing numpy arrays
 Working with 2 dimensional arrays (slicing, indexing, comparison operations)
 Arithmetic operations on 2 dimensional arrays
Iterating 2 dimensional arrays using for loops
 Operations on 2 dimensional arrays (stacking and splitting , vstack, hstack, vsplit and hsplit
operations)

Lab Exercises (2 hrs)

DAY - 2

 Explain the Data Structures with examples?

 How to create, manipulate the data frames?
 Reading data from various sources.
 Indexing, sorting, rank.

Lab Exercises (2 hours)

DAY -3

 Merge, join, concatenate.

 Reshaping,pivoting,duplicating,mapping,replacing
 Summary statistics (Mean, Median, Mode, Skewness, Kurtosis)

Lab Exercises (2 hours)

DAY-4

 What are the Visualization libraries?

 Overview on seaborn and matplotlib packages
 What are the Various plot using this visualization libraries?
 Distribution plots – Histogram, frequency polygon
 Representing data using charts – bar chart, pie chart
 Checking for data anomalies and outliers – box plots
 Association between variables – correlation heatmap, scatter plots, pairplots

Lab Exercises (2 hours)

 EXPLORATORY DATA ANALYSIS
DAY - 1
 What is EDA?
 All EDA Plots?
 Getting data from multiple sources (csv, text, url etc.,)?
 Types of variables – Categorical and Continuous?
 Measures of central tendency - Mean, median, mode?
 Measures of dispersion – Quartiles, percentiles, Standard deviation, variance, coefficient of
variation?
 Coefficient of correlations, Skewness and kurtosis?
 Understand different types of data and measurements and dtype conversions?

Lab Exercises (2 hours)

DAY - 2
Univariate
 Visualising and understanding the data at hand, Summary statistics Moments, distributions
 Data transformation – z-score, normalisation,
 Label Encoding, One hot encoding,Replacing data.
 Coefficient of correlations, Skewness and kurtosis.
 Scaling Vs Normalization

Lab Exercises (2 hours)

DAY - 3
Bivariate
 Feature to feature relationships
 Correlation and Frequency tables
 Seasonality and looking at trended data
 Multi variate analysis

Lab Exercises (2 hours)

DAY - 4
Wrangling
 Various ways of treating missing value’s / Missing value Treatment?
 Various ways of outlier treatment?
 Data Imbalance treatment
 Feature engineering, Introduction to Test and Train

Lab Exercises (2 hours)

 SQL
DAY-1
 Introduction to Database
 Types of Data models
 Database Operations
 Advantages of DBMS
 Introduction to Normalization.
 Need for Normalization/Denormalization.
 Normalization:
- 1 Normal Formal
ST

- 2 Normal Formal
ND

- 3 Normal Formal
RD

 DDL- Understand how to create different database objects.

 DDL to Create and Manage Tables

 DDL Expressions - CREATE, ALTER, DROP, TRUNCATE, COMMENT, RENAME.

 Types of Constraints.

 DML- Describe how to manipulate data.

 Explain how to retrieve data from tables using Filter’s

 In Class Lab Exercise: - Sql Queries Using select, where, order by Expressions.
 Take Home Lab Exercise :- Sql Queries Using select, where, order by Expressions.

DAY -2

 Functions in SQL.
 Types of Single Row Functions
 Explain how to use single-row functions in SQL
 Explain how to use group functions in SQL
 In Class Lab Exercise: - Sql Single-Row Functions, Group Functions Expressions.

 Take Home Lab Exercise: - Sql Single-Row Functions, Group Functions Expressions.
DAY-3

 Explain how to retrieve data from multiple tables using joins

 Apply set operations on tables
 Views , Advantages of Views , Creating a View , Creating a Complex View ,
Rules For Performing DML Operations on a View.
 Using the Check option clause , Indexes , Creating an Index.

 In class Lab Exercises: - SQL JOINs

 Take Home Lab Exercises: - SQL JOINs

DAY-4

 Show How to use Sub-Queries.

 Group Functions in Subqueries.

 Subqueries Within Subqueries using any clause, using ‘ALL’ clause, ‘IN’, ‘NOT IN ‘ clause ,

multiple-columns in subquery, ‘Null’ Values in Subquery, Correlated Subquery, Exists

Operator.

 Sub-queries with aggregate functions.

 Sub-queries with conditional logic.

 Joining sub-queries.

 SQL windows functions

 SQL String functions to clean data

 In class Lab Exercises : -Subqueries .

 Take Home Lab Exercises : -Subqueries .

 Stats for ML + Linear Regression
DAY - 1:
 Data and measurement
 Types of data measurement scales – Discrete, Continuous and categorical data.
 Sample vs. Population
 Description of Data
 Measures of central tendency – mean, median and mode
 Measures of dispersion –
 Quartiles, Standard deviation and Variance.
 Graphically examining dispersion – Histogram and frequency
polygon.
 Measuring association between variables – Covariance and coefficient of correlation
 Summarizing data – The 5-point summary
 Intro to basics of probability
 Definition of probability - Mathematical vs statistical interpretation
 Multiple events - Addition and multiplication in probability
 Probability distribution - Enumeration of all possible outcomes
 Introduction to the Normal distribution
 Sampling distribution and population parameters
 Comparing sample and population - The concept of testing hypotheses
 Sampling error
 Defining a confidence interval

DAY - 2:
 Introduction to hypothesis testing
 Defining a null and alternate hypothesis
 Types of alternate hypothesis - One tail vs two tail test
 Type 1 and type 2 error
 Hypothesis testing applications using the z test
 Interpreting test results
 P-value vs confidence interval approach
 Testing hypothesis for sample - the t distribution
 Testing joint hypothesis - One-way ANOVA and F-test
DAY - 3:
 Examining causal relationship between variables
 Introduction to the concept of regression
 Review - Equation of a straight line
 Visualizing regression line as an average line
 Error variance and minimization
 The methodology behind OLS estimation
 Linear regression with a single independent variable
 Fitting a regression line to a dataset
 Sample vs Population regression function
 Hypothesis testing in the context of regression
 Interpreting ANOVA for a regression model
DAY - 4:
 Extending linear regression to more than one independent variable
 Assumptions in linear regression –
- Multicollinearity - effects and detection, VIF
- Autocorrelation - Why is it a problem, Durbin-Watson test
- Heteroscedasticity - What to do when error terms are unequally distributed
 Hypothesis testing in Multiple regression
 Testing individual coefficients vs joint hypothesis
 ANOVA for multiple regression - F test
 The coefficient of determination – R^2
 The adjusted R^2
 Interpreting summary results after fitting a regression model
 Regression + Feature Engineering
Day 1: -
 Introduction to machine learning
 Supervised vs unsupervised learning
 Looking at regression through the perspective of machine learning
 Accuracy scores as a metric of model performance
 Measuring the importance of individual variables in a regression model
 Review - testing for individual significance vs joint significance
 Using the adjusted R^2 to compare model with different number of independent variables
 Approaches to feature selection
 Forward and backward selection
 Parameter tuning and Model evaluation

Day 2: -
 Extending linear regression
 Data transformations and normalization
 Log transformation of dependent and independent variables
 Case study: -
 Dealing with categorical independent variables
 One hot encoding vs dummy variable regression
 Case study on linear regression

Day 3: -
 Modelling probabilistic dependent variables
 The sigmoid function and odds ratio
 The concept of logit
 The failure of OLS in estimating parameters for a logistic regression
 Introduction to the concept of Maximum likelihood estimation
 Advantages of the maximum likelihood approach
 Modelling a logistic regression problem with a case study
 Making predictions and evaluating parameters

Day 4: -
 Extending the logistic model to multi-class predictions
 The one vs all approach
 Multiclass classification
 Case study on Logistic regression –
 Binary classification and Multiclass one vs all
 SUPERVISED LEARNING CLASSIFICATION
DAY - 1:

 Classification Problems – Examples.

 Binary classification vs Multi class classification.
 Decision trees – Simple decision trees. Visualizing decision trees and nodes and splits.
 Working of the Decision tree algorithm.
 Importance and usage of Entropy and Gini index.
 Manually calculating entropy using gini formula and working out how to split decision nodes
 Evaluating decision tree models.
 Accuracy metrics – precision, recall and confusion matrix
 Interpretation for accuracy metric.
 Building a a robust decision tree model. k-fold cross validation - Advantages against simple
train test split.

DAY - 2:
 CART - Extending decision trees to regressing problems.
 Advantages of using CART.
 Multiple decision trees - Introduction to the Random forests algorithm.
 The KNN Algorithm and its workings
 Choosing the optimum number of neighbour’s. Iterating over various values for K .
 Measures of distance used in KNN.
 Evaluating a KNN Model - Accuracy metrics and k-fold cross validation.

DAY - 3:

 The Bayes theorem. Prior probability.

 The Gaussian NAÏVE’S BAYES Classifier.
 Assumptions of the Naive Bayes Classifier.
 Functioning of the Naïve’s bayes algorithm.
 Evaluating the model - Precision, Recall, Accuracy metrics and k-fold cross validation
 ROC Curve and AUC for binary classification for Naive Bayes.
 Extending Bayesian Classification for multiclass classification.

DAY - 4:

 Using the 2 classification case studies and working through all the classication algorithms i.e.,
CART, KNN and Naive Bayes Algorithms sequentially.
 Comparing and evaluating various classification algorithms.
 Feature selection for classification algorithms.
 Advantages and pitfalls of each algorithm. When to use which type of model?
 UNSUPERVISED LEARNING
DAY - 1:

 What is Unsupervised learning?

 The two major Unsupervised Learning problems - Dimensionality reduction and clustering.
 Clustering algorithms.
 The different approaches to clustering – Heirarchical and K means clustering.
 Heirarchical clustering - The concept of agglomerative and divisive clustering.
 Agglomerative Clustering – Working of the basic algorithms.
 Distance matrix - Interpreting dendograms.
 Choosing the threshold to determine the optimum number of clusters.
 Case Study on Agglomerative clustering

DAY - 2:

 The K-means algorithm.

 Measures of distance – Euclidean, Manhattan and Minowski distance.
 The concept of within cluster sums of squares.
 Using the elbow plot to select optimum number of cluster’s.
 Case study on k-means clustering.
 Comparison of k means and agglomerative approaches to clustering.

DAY - 3:

 Noise in the data and dimensional reduction.

 Capturing Variance - The concept of a principal components.
 Assumptions in using PCA.
 The working of the PCA algorithm.
 Eigen vectors and orthogonality of principal components.
 What is complexity curve?
 Advantages of using PCA.
 Bulid a model using Principal components and comparing with normal model. What is the
difference?

DAY - 4:

 Putting it all together.

 The relationship between unsupervised and supervised learning.
 Case study on Dimensionality reduction followed by a supervised learning model.
 Case study on Clustering followed by classification model.
 TABLEAU
Below Topices to be covered with 4 day-wise breakup.
 Introduction to Visualization, Rules of Visualization
 Data Types, Sources, Connections, Loading, Reshaping
 Data Aggregation
 Working with Continuous and Discrete Data
 Using Filters
 Using Calculated Fields and parameters
 Creating Tables and Charts
 Building Dash Boards and story boards
 Sharing Your Work and Publishing for wider audience

 Importing data from different sources

 Examining imported data
 Linking multiple data tabs/tables
 Understanding ‘Show Me’ features
 Understanding how data is aggregated by Tableau by default – how and when to change them
 Examine a Problem Statement
 Design a Dash Board with appropriate charts
 Build different charts
 Build Story Board
 Publish