Data Science With Advanced AI

2.1 INNER JOIN

2.2 LEFT JOIN

2.3 RIGHT JOIN

2.4 FULL OUTER

2.5 JOIN SELF JOIN

Hands-On:

• Get a list of all employees and their department names using INNER JOIN.
• Create a SELF JOIN to find pairs of employees who work in the same department.
•  Use a LEFT JOIN to find employees and their departments, including those without a department.

3.1 Data Manipulation: INSERT, UPDATE, DELETE Statements

3.2 Transactions and ACID Properties (BEGIN TRANSACTION, COMMIT, ROLLBACK)

3.3 String Functions: CONCAT(), SUBSTRING(), TRIM()

3.4 Date Functions: DATE(), DATEDIFF()

3.5 Mathematical Functions: ROUND(), MOD()

3.6 Conditional Logic (CASE WHEN)

Hands-On:

• Insert a new employee into the employees table.
• Update the salary of all employees in the “Sales” department by 10%.
• Write a transaction to increase salaries and roll back if an error occurs.
• Concatenate first and last names of employees into full_name.
• Calculate the number of days an employee has worked since the hire date.
• Use CASE WHEN to categorize employees’ salaries as ‘Low’, ‘Medium’, or ‘High’.

4.1 Window Functions: ROW_NUMBER(), RANK(), DENSE_RANK()

4.2 Lead and Lag Analysis (LEAD(), LAG())

4.3 OVER() Clause

4.4 Data Definition: CREATE, ALTER, DROP TABLE

4.5 Constraints: PRIMARY KEY, FOREIGN KEY, NOT NULL, UNIQUE

4.6 Indexes: Creating and Managing

Hands-On:

• Use ROW_NUMBER() to assign a unique rank to each employee based on salary.
• Write a query using LEAD() to compare each employee’s salary with the  next employee’s salary.
• Rank employees within  their departments based on salary using RANK().
• Create a projects table with appropriate data types and constraints.
• Add a NOT NULL constraint to the project_name column in the projects table.
• Create an index on the employee_name column to speed up search queries.

5.1 Views and Their Uses

5.2 Stored Procedures

5.3 User-Defined Functions

5.4Query Optimization (EXPLAIN)

5.5 Efficient Use of Indexes

5.6 Data Security (GRANT, REVOKE)

Hands-On:

• Create a view called high_earners that shows employees earning more than $75,000.
• Write a stored procedure to return the average salary for a given department.
• Create a function to calculate an employee’s yearly bonus based on their salary.
• Analyze the performance of a slow-running query using EXPLAIN and optimize it.
• Partition the orders table by order_date for performance improvement
• Grant SELECT and UPDATE privileges to a user on the sales table, then revoke DELETE privileges.

6.1 Python Basics: Syntax, Variables, Data Types

6.2 Input and Output (print(), input())

6.3 Basic Operators and Expressions

6.4 Conditional Statements (if, elif, else)

6.5 Loops (for, while)

6.6 Data Structures: Lists, Tuples, Sets, Dictionaries

6.7 List and Dictionary Comprehensions

6.8 Iterating over Data Structures

Hands-On:

• Write a Python script to calculate the total cost of a product based on user input for price and quantity.
• Create a program to check if a number is even or odd.
• Create a list of customer names and sort them alphabetically.
• Create a dictionary mapping products to prices and retrieve a product’s price based on user input.

7.1 File Handling: Reading and Writing Files (Text and CSV)

7.2 Parsing CSV Files with the csv Module

7.3 Error Handling with Try/Except Blocks

7.4 Introduction to Numpy Arrays and Array Operations

7.5 Reshaping, Indexing, and Slicing Numpy Arrays

7.6 Applying Mathematical Functions and Aggregations to Numpy Arrays

7.7 Working with Multi-Dimensional Data

Hands-On:

• Write a script that reads customer data from a text file and prints it to the console.
• Write a program to read sales data from a CSV file, calculate total sales per product, and write the results to a new CSV file.
• Create a Numpy array from a dataset and perform basic statistical operations (sum, mean, variance).
• Reshape a multi-dimensional array and apply transformations (e.g., logarithmic, square root).

8.1 Introduction to Pandas DataFrames and Series

8.2 Loading Data from CSV, Excel, or SQL Databases

8.3 Basic Data Frame Operations (Selection, Filtering, Sorting, Grouping)

8.4 Handling Missing Data

8.5 Data Cleaning Techniques (Dealing with Missing Data, Duplicates, Outliers)

8.6 Data Transformation: Renaming Columns, Changing Data Types, Applying Functions   

8.7 Handling Dates and Times with pd.to_datetime()

8.8 Combining and Merging DataFrames

Hands-On:

• Load a sales dataset into a Pandas DataFrame and perform operations such as filtering for specific products, sorting by sales, and handling missing values.
• Write a script to group sales data by region and calculate total sales for each region.
• Clean a dataset by removing duplicates, filling missing values, and transforming date columns into a standard format.
• Merge two DataFrames containing customer and sales data, then calculate total sales per customer.

9.1 Descriptive Statistics with Pandas (mean(), median(), mode(), std())

9.2 Data Visualization using Matplotlib and Seaborn (Bar Plots, Histograms, Box Plots)

9.3 Correlation Analysis and Pairwise Relationships

9.4 Detecting and Visualizing Outliers

9.5 Advanced Visualization Techniques with Matplotlib and Seaborn

9.6 Line Plots, Scatter Plots, and Pie Charts

9.7 Customizing Plots (Titles, Labels, Legends, Colors)

9.8 Creating Subplots and Multi-Panel Figures

Hands-On:

• Perform an EDA on a dataset, calculate basic statistics, and visualize relationships using bar plots and histograms.
• Visualize correlations between numeric variables in a dataset and create a heatmap.
• Create a multi-panel visualization showing sales trends over time and sales distribution by product.
• Generate a pie chart to show the market share of different products.

10.1 Introduction to Databases with SQLAlchemy

10.2 Writing SQL Queries from Python Using SQLAlchemy ORM

10.3 Loading Data from SQL Databases into Pandas

10.4 Writing Data Back to SQL Databases from Pandas

10.5 Introduction to Views, Stored Procedures, Functions, and Performance Optimization in Databases

Hands-On:

• Write a Python script that connects to a SQL database, extracts sales data, and loads it into a Pandas DataFrame for analysis.
• Perform analysis on the data (e.g., find the top customers by revenue) and write the results back to the–

11.1 Univariate Analysis

11.2 Bi Variate Analysis

11.3 Multi Variate Analysis

11.4 Central Tendencies Mean, Median, Mode, Variance Standard Deviation,

11.5 Distribution Probability Distribution

11.6 Normal Distribution

11.7 Hypothesis Testing

11.8 Confidence Interval

11.9 Correlation Vs Covariance

11.10 Central Limit Theorm P value/ Alpha

11.11 Intro to Advance/ Inferential Statistics Hypothesis Testing

11.12 T test Z test Chi Square Anova

12.1 What is Machine Learning?

12.2 Why Machine Learning?

12.3 Limitations of Machine Learning

12.4 Prediction Modelling Linear Regression

12.5 Sales Predictions and Forecasting

12.6 Classification using Logistics Regression

12.7 Overview of Supervised and Unsupervised Learning

Hands-On:

• Identify examples of supervised and unsupervised learning from a given dataset.
• Churn Analysis
• Fraud Analysis

13.1 Introduction to scikit-learn

13.2 Core Components of Scikit-learn for Machine learning

13.3 Understanding Regression Problems

13.4 Define regression problems

13.5 Dependent and independent variables, and their roles.

13.6 Train a Regression model using scikit-learn.

13.7 Techniques for assessing model performance

13.8 Techniques for improving model performance, such as feature scaling and polynomial regression.

13.9 Train a K-Means model on the dataset and visualize the clusters formed.

Hands-On:

• Install scikit-learn and load a sample dataset. Practice importing and understanding datasets within scikit-learn.Load a dataset with continuous target variables and identify the dependent and independent variables.
• Train a simple linear regression model on a sample dataset, such as predicting house prices
• Evaluate the trained regression model using MSE and R² on both the training and test datasets
• Improve the regression model by scaling features and applying polynomial regression; compare model results.

14.1 Linear Regression

14.2 Logistics Regression

14.3 Decision Trees

14.4 Random Forest

14.5 Support Vector Machine (SVM)

14.6 K-Nearest Neighbors (KNN)

14.7 Time Series Forecasting

Hands-On:

• Classifying Heart Disease Risk with Logistics Regression
• Customer Churn Prediction with Decision Tree
• Predicting Stock Prices using Time Series Forcasting

15.1 K-Means Clustering

15.2 Dimensionality Reduction

15.3 Linear Discriminant Analysis (LDA)

15.4 Principal Component Analysis (PCA)

Hands-On:

• Linear Discriminant Analysis (LDA) Exercise: Facial Recognition
• Principal Component Analysis (PCA) Exercise: Image Compression
• Feature Reduction and Classification in a High-Dimensional Dataset

16.1 Overview of Deep Learning and Neural Networks

16.2 Difference between DS VS ML VS AI

16.3 Setting up TensorFlow, Keras, and PyTorch environments

16.4 Introduction to Tensors, Computation Graphs, and Autograd

16.5 Basics of building and training models in TensorFlow and PyTorch

Hands-On:

• Set up TensorFlow, Keras, and PyTorch environments.
• Create a simple neural network in both TensorFlow and PyTorch.
• Implement basic tensor operations (addition, multiplication, reshaping).
• Visualize computation graphs in TensorFlow.

17.1 Artificial Neural Networks (ANN)

17.2 Convolutional Neural Networks (CNN)

17.3 Recurrent Neural Networks (RNN)

17.4 Introduction to Transfer Learning

Hands-On:

• Build and train a basic ANN for a simple classification task (e.g., MNIST dataset).
• Implement a CNN for image recognition using a standard dataset (e.g., CIFAR-10).
• Train an RNN model to predict a sequence (e.g., text generation or temperature prediction).
• Apply transfer learning to a CNN model for a custom image dataset

18.1 Long Short-Term Memory Networks (LSTM)

18.1 Graph Neural Networks (GNN)

18.1 Advanced Recurrent Architectures: Bidirectional RNN, GRU

18.1 Hyperparameter Tuning and Model Optimization

Hands-On:

• Train an LSTM model on a time-series dataset (e.g., stock prices or weather data).
• Implement a GNN to classify nodes in a graph dataset.
• Experiment with Bidirectional RNNs on text data.
• Perform hyperparameter tuning on a CNN model to optimize accuracy.

19.1 Introduction to NLP concepts and text preprocessing (tokenization, stemming)

19.2 Using NLTK for basic NLP tasks

19.3 Text classification using deep learning (RNN, LSTM)

19.4 Sentiment analysis and text generation

Hands-On:

• Preprocess a text dataset using NLTK (e.g., tokenization, stop word removal).
• Build a text classification model (e.g., spam detection) using an RNN or LSTM.
• Train a sentiment analysis model on movie reviews or Twitter data.
• Experiment with a text generation model for generating sentences or paragraphs.

20.1 Advanced CNNs and image processing techniques

20.2 Object detection and YOLO (You Only Look Once) for real-time detection

20.3 Transfer Learning with pre-trained image models (e.g., ResNet, Inception)

20.4 Fine-tuning models for specific tasks

Hands-On:

• Train a CNN for a custom image recognition task (e.g., animal classification).
• Apply YOLO for real-time object detection in images or videos.
• Experiment with transfer learning using a pre-trained model on a new dataset.
• Fine-tune the YOLO model on a specific category of objects (e.g., cars or animals).

21.1 Leveraging Gen AI for Bussiness Analytics

21.2 ML Foundation for Open AI

21.3 Bussiness Landscape and Overview

21.4 Prompt Engineering w/o Code

21.5 Gen AI Models – BERT, GPT, DALL-E and other Framework (Hugging Face Transformers, Open AI)

21.6 Knowledge, Training and Fine tuning of GenAI models for Industries

22.1 Introduction to MLOps

22.2 MLOps vs. DevOps

22.3 SDLC Basics

22.4 What is Cloud Computing

22.5 AWS Introduction

22.6 S3

22.7 Waterfall vs AGILE vs DevOps vs MLOps

22.8 MLOps Phases

22.9 Model Packaging

23.1 Git Essentials

23.2 Configuring Git

23.3 Branching

23.4 Git Workflow

23.5 Repo

23.6 Git Commands

23.7 Tracking and managing changes to code

23.8 Source Code Management

23.9 Tracking and Saving Changes in Files

24.1 Introduction to CI/CD

24.2 CI/CD Challenges

24.3 CI/CD Implementation in ML

24.4 Popular DevOps Tools

24.5 AWS CodeCommit

24.6 AWS CodePipeline

24.7 AWS CodeBuild

24.8 AWS CodeDeploy

25.1 Docker Architecture

25.2 Docker for Machine Learning

25.3 Continuous Deployment

25.4 Writing a Dockerfile to Create an Image

25.5 Installing Docker Compose

25.6 Configuring Local Registry

25.7 Container Orchestration

25.8 Application Deployment

25.9 Kubernetes Core Concepts