AI Machine Learning Deep Learning Outline

The aim for the team of staff responsible for delivery of the advanced IT curriculum is to provide knowledge and develop skills related to the IT. The course will allow participants to gain a comprehensive understanding of all the aspects. It will also develop the participant’s ability to act in a professional and responsible manner.

Teaching staff will provide the technical knowledge and abilities required to solve tasks and problems that are goal-oriented. They will use participant-centered, practically oriented methods. They will also develop a program of practical assessment that reflects the learning outcomes stated in the curriculum. Trainees of the IT curriculum will also develop their willingness and ability as individuals to clarify issues, as well as think through and assess development opportunities.

Teaching staff will also support trainees in developing characteristics such as self-reliance, reliability, responsibility, a sense of duty and a willingness and ability to criticize and accept criticism well and to adapt their future behavior accordingly.

Teaching staff also use the IT curriculum to address the development of professional competence. Trainees will acquire the ability to work in a professional environment. By the end of this course, the trainees should gain the following competencies:

Understanding of core concepts of artificial intelligence and machine learning

State of the art machine learning techniques Hands-on exposure to exploratory data analysis Practical exposure to model design, evaluation

Familiarity with tools and libraries such as scikit learn, pandas numpy, tensorflow, pytorch and keras

By the end of this course, the trainees should gain the following competencies:

Understanding of core concepts of artificial intelligence and machine learning

State of the art machine learning techniques Hands-on exposure to exploratory data analysis Practical exposure to model design, evaluation

Familiarity with tools and libraries such as scikit learn, pandas numpy, tensorflow, pytorch and keras

After the specialization in NLP, you will be comfortable using TensorFlow pipelines for NLP at the end of the course. Moreover, You will learn to build your own models which will extract information from textual data. You will learn text processing fundamentals, including text normalization, stemming and lemmatization. You will learn about different evaluation metrics for models trained for NLP tasks. You will learn to make a part of speech (POS) tagging model. You will learn about named entity recognition. You will learn advanced techniques including word embeddings, deep learning (DL) techniques. You will learn how to deploy a NLP model

Moreover, you will learn not only all these skills but also learn to use Microsoft Azure API for Machine and Deep Learning for numerical, image and text data.

2.    Afiniti

3.    Addo.ai

4.    Arbisoft

5.    I2c

6.    Xavor

7.    Fiverivers Technologies

8.    Confiz

9.    Crossover

10.  NetSol

11.  Research institutes

12.  All Private Institutes who have an ML department

is the very nature of the subject that makes its applications infinite,

·         AI Associate Engineer

·         Machine Learning associate analyst

·         Assistant Data Analyst

·         Research Assistant

/ Reference Material

·         Learn Linux Shell Scripting – Fundamentals of Bash 4.4

[Sebastiaan Tammer – Packt Publishing Ltd.]

·         Sams Teach Yourself Shell Programming in 24 Hours

[Second Edition , Sams Publishing]

·         Applied Data Science – (Chapter 01) [Ian Langmore & Daniel Krasner]

·         Linux Tutorial – Basic Command Line https://www.youtube.com/watch?v=cBokz0LTizk

Python:

·         Learning Python – 2nd Edition (Ch:12: OOP in Python)

[B. Nagesh Rao, CyberPlus Infotech Pvt. Ltd.]

·         Python for Everybody

[Dr. Charles R. Severance]

·         Python: A Simple Tutorial

[Matt Huenerfauth, University of Pennsulvania, USA]

·         Smarter Way to Learn Python [Mark Mayers]

·         A Python Book: Beginning Python, Advanced Python, and Python Exercises

[Dave Kuhlman]

·         Mastering Object-Oriented Python

[Second Edition, Steven F. Lott, Pack Publishing Ltd.]

·         Python Official Documentation https://docs.python.org/3/

Descriptive Statistics and Probability:

·         Probability for Machine Learning [Jason Brownlee]

·         Making Sense of Data: A Practical Guide to Exploratory Data Analysis and Data Mining (Ch: 02) [Second Edition, Glenn J. Myatt & Wayne P. Johnson, WILEY]

·         Practical Statistics for Data Scientists

[Second Edition, Peter Bruce, Andrew Bruce, and Peter Gedeck, O’REILLY]

·         Numpy

o   Python for Data Analysis

(Ch:04, Appendix A: Advanced Numpy) [Second Edition, Wes McKinney, O’REILLY]

o   Numpy Official Documentation https://numpy.org/doc/1.24/

·         Pandas

o   Pandas 1.x Cookbook

[Second Edition, Matt Harrison & Theodore Petrou, Pack Publishing Ltd.]

o   Python for Data Analysis (Ch:05, 07, 10, 12)

[Second Edition, Wes McKinney, O’REILLY]

o   Hands-on Exploratory Data Analysis with Python

(Ch: 04, 06)

[Suresh Kumar Mukhiya & Usman Ahmed, Pack Publishing Ltd.]

o   Pandas Official Documentation https://pandas.pydata.org/docs/

·         Matplotlib

o   Pandas 1.x Cookbook (Ch:13)

[Second Edition, Matt Harrison & Theodore Petrou, Pack Publishing Ltd.]

o   Hands-on Exploratory Data Analysis with Python

(Ch: 04, 06)

[Suresh Kumar Mukhiya & Usman Ahmed, Pack Publishing Ltd.]

o   Matplotlib Official Documentation https://matplotlib.org/stable/index.html

·         Seaborn

o   Pandas 1.x Cookbook (Ch:13)

[Second Edition, Matt Harrison & Theodore Petrou, Pack Publishing Ltd.]

o   Python for Data Analysis (Ch:09)

[Second Edition, Wes McKinney, O’REILLY]

o   Seaborn Official Documentation https://seaborn.pydata.org/

·         Machine Learning by Andrew NG (Also available freely on Youtube) https://www.coursera.org/collections/machine- learning

·         Machine Learning: An Algorithmic Perspective [Second Edition, Stephen Marsland, CRC Press]

·         Hands-On Machine Learning with Scikit-Learn,

Keras, and TensorFlow

[Third Edition, Aurélien Géron, O’REILLY]

·         XGBoost with Python [Jason Brownlee]

·         Learn TensorFlow 2.0

[Pramod Singh & Avinash Manure, Apress]

Natural Language Processing:

·         Speech and Language Processing

·         [Third Edition, Dan Jurafsky, James H. Martin]

·         Deep Learning for Natural Language Processing [Jason Brownlee]

·         Natural Language Processing Cookbook

[Krishna Bhavsar, Naresh Kumar, & Pratap Dangeti, Pack Publishing Ltd.]

Deep Learning:

·         Deep Learning by Andrew NG (Also available freely on Youtube)

·         https://www.coursera.org/learn/neural-networks- deep-learning

·         Deep Learning with Python [Jason Brownlee]

·         Deep Learning for Time Series Forecasting [Jason Brownlee]

·         Long Short-Term Memory Networks with Python [Jason Brownlee]

·         [Jason Brownlee]

·         Dive into Deep Learning

[Aston Zhang, Zachary C. Lipton, Mu Li, and Alexander J. Smola]

Microsoft Azure Machine Learning:

·         Mastering Azure Machine Learning: Execute Large- Scale End-to-end Machine Learning with Azure [Second Edition, Christopher Korner and Marcel Alsdorf, Packt Publishing Ltd.]

·         Microsoft Azure AI Fundamentals Training

·         Microsoft Azure AI Associate Training https://learn.microsoft.com/en- us/training/paths/prepare-teach-ai-102-microsoft- design-implement-azure/

·         Microsoft Learn for Educators Program https://learn.microsoft.com/en-us/training/educator- center/programs/msle/

Software Download:

·         Anaconda https://www.anaconda.com/

·         VSCode https://code.visualstudio.com/

·         PyCharm (Community Edition) https://www.jetbrains.com/pycharm/

·         PyTorch

https://pytorch.org/get-started/locally/

·         TensorFlow 2.0 https://www.tensorflow.org/install

Linux Shell Scripting Fundamentals

Python Fundamentals

·         Motivational Lecture (For further detail please see Page No: 3& 4)

·         Task 1

·         Task 2

·         Task 3-25

Details may be seen at  Annexure-I

·         Job market

·         Course Applications

·         Work ethics

·         Survey of career opportunities

·         Survey of industry requirements for each career path

·​

3, 4

·   Basic Commands: pwd, cd, ls, cat, sudo, man, redirection, mkdir, rm, rmdir, cp, mv

2

3

4

·   Environment Variables

1

·   Numbers, Booleans, Strings

·   Operators, variables and keywords

# 2,3

4

·   Comments

1 & 2

·   Lists

·   Tuples

3 & 4

·   Sets

1 & 2

·   If, elif, and else statements

·   Break, continue, and pass statements

·   Conditional (Ternary) Expression

3 & 4

·   Nested loops

·   List comprehension

·   Iterators and Iterables

·​

Implementation of OOP Principals in Python

Descriptive Statistics and Probability

27

·         Task 49-

51

Details may be seen at  Annexure-I

·         Functions and variable scope

·         Lambda expression

·         Map and Filter

·         Inner/Nested functions

4

·         Exception Handling

·         Instance Variables and Methods

·         Class Variables and Functions

·         Constructors and Destructors

·         Multilevel Inheritance

·         Hierarchical Inheritance

·         Multiple Inheritance, Method Resolution Order

·         Name Mangling

·         Inner/Nested Class

·         Association, Aggregation, Composition

1

·         Keyword Arguments

·         Quantitative data, Numerical, Continuous, and Discrete variables

3-4

·         Mean, Mode, Median

·         Variance, Standard deviation

·         Co-efficient of variation, skewness and kurtosis

·         Z-Score, Percentile, Quartile

Python Support Libraries for Exploratory Data Analysis

–     NUMPY

48

Details may be seen at  Annexure-I

2

·​

·         Bayesian Probability

·         Array Attributes and Methods (reshape, max,

min, argmax, argmin, shape, dtype, size, ndim)

·         Operations on Arrays (copying, append and

Insert, Sorting, Removing/Deleting, Combining/Concatenating, Splitting)

·         NumPy Indexing and Selection (Indexing a 2D array, Logical Selection)

·         Broadcasting

·         Arthmetic Operations (Add, Subtract, Multiply, Divide, Exponentiation)

·         Universal Array Functions (sqrt, exp, max,

–     Pandas

1

·         Selection and Indexing (rows, columns, conditional selection, selection of subset of rows and columns, index setting, etc)

ordering, null value check, value replacement, dropping rows and columns, etc)

4

–       Pandas

–       Seaborn

·         Task 28-48

Details may be

outer, right and left joins)

·         Discretization and Binning

·         Operations on DataFrames

·         Data output/saving

·         Pandas for Plotting (area, bar, density, hist,

line, scatter, barh, box, hexbin, kde, and pie plots

Annexure-I

·         distplot

·         jointplot (pairplot, rugplot, kdeplot)

·         factorplot, boxplot, violinplot, stripplot, swarmplot, barplot, countplot

·         Heatmap

modeling, testing and evaluation, and deployment)

·         Regression and classification problems

·         Components of supervised machine learning (labeled data, hypothesis, cost function,

optimizer)

# 3,4

·         Without Vectorization

Details may be seen at  Annexure-I

2,3,4

4

4

4

4

·         Task 53-

55

Details may be

2

·         NLP curves and future directions

·         Introduction to NLTK/SpaCy

·         Noise removal (stopwords, punctuation, etc)

·         Word segmentation

·         Stemming

·         Text normalization

·         Regular expression for string parsing

Annexure-I

·         NER tagging

·         Chunking and Chinking

·         Lemmatization

·         WordNet

·         Feature Selection and Extraction

·         Document Similarity

·         Classification and Regression

4

# 3-4

Build Your CV – Mid-term Exam

Deep Learning I

·         Task 56-64

Details may be seen at  Annexure-I

4

·         Forward and backward passes

·         Nonlinearity: Activation functions

·         Cross-Entropy

·         Computational graph and Backpropagation

·         Vanishing and exploding gradients

·         Overfitting, underfitting, dropout regularization

4

of choice (TensorFlow, PyTorch, Keras)

·         2D CNN for image classification

# 1-2

4

4

,4

·         Word2vec

·         Continuous BOW

·         Continuous Skip-gram

4

4

4

·         1 to 1

·         1 to Many

4

·         Many to 1

·         Many to Many

Employable Project

/ Assignment

(2 weeks, 11-12) in addition of regular classes.

OR

On job training (2 weeks)

Details may be seen at  Annexure-I

4

4

·         Guidelines to the Trainees for selection of employable project like final year project (FYP).

·         Assignment of Independent project to each Trainee.

·         A project based on trainee’s aptitude and acquired skills.

·         Designed by keeping in view the emerging trends in the local market as well as across the globe.

·         The project idea may be based on entrepreneurship.

·         Leading to the successful employment.

·         The duration of the project will be 2 weeks

·         Ideas may be generated via different sites such as:

https://1000projects.org/

https://nevonprojects.com/ https://www.freestudentprojects.com/

·         Final viva/assessment will be conducted on project assignments.

·         At the end of session, the project will be presented in skills competition.

·         The skill competition will be conducted on zonal, regional and National level.

·         The project will be presented in front of Industrialists for commercialization

·         The best business idea will be placed in NAVTTC business incubation center for commercialization.

OR

·         On job training for 2 weeks:

·         Aims to provide 2 weeks industrial training to the Trainees as part of overall training program

·         Ideal for the manufacturing trades

·         As an alternate to the projects that involve expensive equipment

·         Focuses on increasing Trainee’s motivation, productivity, efficiency and quick learning

approach.

·         Task 65

Details may be

·         Selection the appropriate service for a vision solution

·         Selection the appropriate service for a language analysis solution

4

# 1,2

Services for a speech solution

·         Selection the appropriate Applied AI services

·         Manage account keys

·         Manage authentication for a resource

·         Secure services by using Azure Virtual Networks

·         Plan for a solution that meets Responsible AI principles

·         Create an Azure AI resource

·         Configure diagnostic logging

·         Monitor an Azure AI resource

·         Determine a default endpoint for a service

·         Create a resource by using the Azure portal

·         Integrate Azure AI services into a continuous integration/continuous deployment (CI/CD) pipeline

·         Plan a container deployment

·         Implement prebuilt containers in a connected environment

·         Create a solution that uses Anomaly Detector, part of Cognitive Services

·         Create a solution that uses Azure Content Moderator

·         Create a solution that uses Personalizer, part of Cognitive Services

·         Create a solution that uses Azure Metrics Advisor, part of Azure Applied AI Services

·         Create a solution that uses Azure Immersive Reader, part of Azure Applied AI Services

·         Analyze images

·         Extract text from images

detection by using the Custom Vision service, part of Azure Cognitive Services

·         Task 65

Details may be seen at  Annexure-I

4

·         Analyze text

·         Process speech

·         Translate language

4

·         Create a question answering solution