Program Week Course # Course Name Type Learning Course Week # Weekly Topic Description Concepts Learning Sample

ic Description Concepts Learning Sample Project Project / Case # Projects # Quizzes # Case Studies # Practice
# (Graded / Self- Outcomes (Week) (Week) Outcomes / Case Study Study (Course) (Course) (Course) Exercises
paced) (Course) (Week) (Course) Description (Course)
(Week)
- - Introduction to Self-paced Gain an - - - 1. The - - - 0 0 0 0
Data Science understanding fascinating
and AI of the evolution history of Data
of AI and Data Science and AI
Science over 2. Transforming
time, their Industries
application in through Data
industries, the Science and AI
mathematics 3. The Math
and statistics and Stats
behind them, underlying the
and an technology
overview of the 4. Navigating
life cycle of the Data
building data Science and AI
driven solution Lifecycle
- - Generative AI Self-Paced Get an - - - 1. ChatGPT and - - - 0 0 0 0
overview of Generative AI -
Generative AI, Overview
what ChatGPT 2. ChatGPT -
is and how it Applications
works. delve and Business
into the 3. Breaking
business Down ChatGPT
applications of 4. Limitations
ChatGPT, and and Beyond
an overview of ChatGPT
other 5. Generative AI
generative AI Demonstrations
models/tools
via
demonstrations
- - Pre-Work Self-paced Gain a - - - 1. Python IDE - - - 0 1 0 0
fundamental 2. Introduction
understanding to Python
of the basics of Programming
Python
programming
and build a
strong
foundation of
coding to build
AI applications
- - - Self-paced - - Orientation - - - - - - - - -
- - - Self-paced - - Induction + AI AI is used 1. Application Experience how Hotel Booking Build an AI - - - -
Application across a variety of AI businesses use Cancellation solution for a
Case Study of businesses 2. AI Solution AI with tools & Prediction chain of hotels
to enhance Workflow technologies that will help
operational 3. Measuring involved to them predict
efficiency, the impact of an solve problems the likelihood of
streamline AI solution and gain an a booking
decision- overview of getting
making how an end to cancelled so
processes, gain end AI solution that they can
actionable works take measures
insights from to fill in
data, and foster potential
innovation, This vacancies and
module will reduce revenue
cover an end- loss
to-end case
study to
illustrate the
different stages
of
implementing
an AI solution
for a business
problem.
1 1 Introduction to Graded Read, explore, 1 Python Python is a 1. Variables Learn about the CredPay Cred-Pay is a 1 14 7 0
Python manipulate, and Programming widely used, and Datatypes fundamentals consultation
visualize data high-level, 2. Data of Python firm that
to tell stories, interpreted Structures programming partners with
solve business programming 3. Conditional (variables, data banks and
problems, and language, and Looping structures, checks if their
deliver having a Statements conditional and customers are
actionable simple, easy- 4. Functions looping eligible for a
insights and to-learn syntax statements, credit card.
business that highlights functions) They are a
recommendatio code startup just at
ns by readability. This the roots of
performing module will starting their
exploratory cover the business. They
data analysis fundamentals have partnered
using some of of Python with a few
the most widely programming banks and are
used Python and taking the currently
packages first steps in collecting data
organizing data for credit card
with Python. applications.

You have been

hired as a Data
Scientist to
handle and
organize the
data. You are
responsible to
organize the
data so that it
will be easily
accessible and
to help the
company
predict if the
application can
be accepted for
a credit card or
not.
2 1 Introduction to Graded Read, explore, 2 Python for Data NumPy is a 1. NumPy Learn about MovieLens MovieLens is a
Python manipulate, and Science Python package arrays and two of the most company in the
visualize data for functions commonly used internet and
to tell stories, mathematical 2. Accessing libraries entertainment
solve business and scientific and modifying (NumPy and domain
problems, and computing and NumPy arrays Pandas) used in providing an
deliver involves 3. Saving and Data Science online database
actionable working with loading NumPy for reading and of information
insights and arrays and arrays manipulating related to films,
business matrices. 4. Pandas data television
recommendatio Pandas is a Series series, online
ns by fast, powerful, (Creating, streaming
performing flexible, and Accessing, and content
exploratory simple-to-use Modifying including cast,
data analysis open-source Series) production
using some of library in 5. Pandas crew, trivia,
the most widely Python to DataFrames ratings, fan, and
used Python manipulate and (Creating, critical reviews.
packages analyse data. Accessing, Every year in
This module Modifying, and collaboration
will cover these Combining with a guest
important DataFrames) curator,
libraries and 6. Pandas MovieLens
provide a deep Functions publish their
understanding 7. Saving and annuals based
of how to use loading on a theme
them to explore datasets using providing a
data. Pandas comprehensive
view of a topic.
The company is
planning to
bring out the
"Movie Talkies:
Classic" edition
this year. The
idea is to
explore the
movies that are
a decade old
and deliver a
detailed
analysis.
3 1 Introduction to Graded Read, explore, 3 Python for Matplotlib is a 1. Histogram, Learn about Tips Chef's Kitchen
Python manipulate, and Visualization library to create Boxplots and different visual is one of the
visualize data statically Bar graphs tools which most popular
to tell stories, animated, 2. Line Plot, help in restaurants in
solve business interactive Scatterplot, and summarizing the city of San
problems, and visualisations, Lmplot data better and Diego and acts
deliver whereas 3. Jointplot, how to create as a one-stop
actionable Seaborn is a Violin Plot, and them using destination for
insights and Matplotlib striplot Seaborn, a food lovers. The
business based data 4. Swarm, popular Python polite and
recommendatio visualisation catplot, and library efficient service
ns by library in pairplots provided by the
performing Python. 5. Heatmaps, restaurant staff
exploratory This module Plotly, and often gets them
data analysis will give you a Customizing of tips from the
using some of deep Plots customer. As a
the most widely understanding Data Analyst
used Python of exploring for the
packages data sets using restaurant, you
Matplotlib and have been
Seaborn asked to
analyze the
data provided
to identify the
patterns and
trends in the
revenue and
tips received
from customers
across different
demographies
and come up
with
informative
visualizations to
convey the
insights
obtained from
the analysis.
4 1 Introduction to Graded Read, explore, 4 Exploratory Exploratory 1. Data Learn how to Google Play Zoom Ads is an
Python manipulate, and Data Analysis Data Analysis, overview perform Store advertising
visualize data (Deep Dive) or EDA, is a 2. Univariate Exploratory agency that
to tell stories, process of analysis Data Analysis wants to
solve business examining and 3. (EDA) to extract perform an
problems, and visualizing data Bivariate/Multiv insights from analysis on the
deliver to uncover ariate analysis data data of the
actionable patterns and 4. Missing value google play
insights and extract treatment store. They
business meaningful 5. Outlier need to
recommendatio insights from it detection and understand the
ns by and facilitates treatment trend of
performing storytelling. applications
exploratory This module available on the
data analysis provides a deep google play
using some of insight on how store so that
the most widely to conduct EDA they can decide
used Python using Python to focus on
packages and utilize the promoting
insights advertisements
extracted to on particular
drive business applications
decisions. which are
trending in the
market and can
lead to
maximum
profit. As a
Data Scientist,
you are
required to
gather and
analyze
detailed
information on
apps in the
Google Play
Store in order
to provide
insights on app
features and
the current
state of the
Android app
market.
5 1 Introduction to Self-paced Read, explore, 5 Project Support - - - - -
Python manipulate, and
visualize data
to tell stories,
solve business
problems, and
deliver
actionable
insights and
business
recommendatio
ns by
performing
exploratory
data analysis
using some of
the most widely
used Python
packages
6 1 Introduction to Graded Read, explore, 6 Project - - - FoodHub Order Analyze the
Python manipulate, and Analysis using data of a food
visualize data Python aggregator
to tell stories, company,
solve business answer key
problems, and questions
deliver provided, draw
actionable some
insights and actionable
business insights, and
recommendatio help the
ns by company to
performing improve the
exploratory business
data analysis
using some of
the most widely
used Python
packages
7 2 Applied Graded Perform 1 Inferential Inferential - Experiments, Learn about the Medicon Drug Examine the 1 4 9 3
Statistics statistical Statistics statistics is Events, and fundamentals Testing data provided
analysis using Foundations pivotal in Definition of of probability by a
Python to statistical Probability distributions pharmaceutical
evaluate the analysis and - Introduction and the company
reliability of a decision- to Inferential foundations of regarding a
particular making and Statistics Inferential new batch of
business involves - Introduction Statistics COVID-19
estimate using drawing to Probability vaccine doses
confidence conclusions Distributions and use
intervals and about (Random statistical
test hypotheses populations Variable, analysis to
(assumptions) based on Discrete and check the time
before putting samples. This Continuous of effect of
them into action module will Random doses and
and committing introduce Variables, conduct quality
resources by learners to the Probability assurance
analyzing data common Distributions) checks
distributions probability - Binomial
and performing distributions Distribution
hypothesis and how they - Normal
testing are used to Distribution
make - z-score
statistically-
sound, data-
driven
decisions.
8 2 Applied Graded Perform 2 Estimation and Estimation - Sampling Learn about the Talent Hunt Analyze the
Statistics statistical Hypothesis involves - Central Limit Central Limit Examination student
analysis using Testing determining Theorem Theorem, performance
Python to likely values for - Estimation Estimation, and data of a
evaluate the population - Introduction the key research
reliability of a parameters to Hypothesis concepts of institute
particular from sample Testing (Null Hypothesis conducting
business data, while and Alternative Testing examinations
estimate using hypothesis hypothesis, every year to
confidence testing provides Typ-I and hire people
intervals and a framework for Type-II errors, who can work
test hypotheses drawing alpha, critical on various
(assumptions) conclusions region, p-value) research
before putting from sample - Hypothesis projects across
them into action data to the Formulation different fields
and committing broader and Performing of study, extract
resources by population. This a Hypothesis summary points
analyzing data module covers Test that can be
distributions the important - One-tailed used for
and performing concepts of and Two-tailed marketing
hypothesis central limit Tests purposes, and
testing theorem and - Confidence conduct a
estimation Intervals and hypothesis test
theory that as Hypothesis to check the
vital for Testing validity of a
statistical claim made by
analysis, and a reputed critic
the framework regarding the
for conducting institute
hypothesis
tests.
9 2 Applied Graded Perform 3 Common Hypothesis - Common Learn about Diet Case Study Analyze the
Statistics statistical Statistical Tests tests assess the Statistical Tests various data provided
analysis using validity of a - Test for one commonly used by a health
Python to claim or mean statistical tests company
evaluate the hypothesis - Test for and their regarding a
reliability of a about a equality of implementation market test
particular population means (known in Python with experiment to
business parameter standard business check the
estimate using through deviation) examples effectiveness of
confidence statistical - Test for various diet
intervals and analysis. This equality of plans for
test hypotheses module means (Equal weight loss,
(assumptions) introduces and unknown and conduct
before putting learners to the std dev) hypothesis
them into action most commonly - Test for tests to find
and committing used equality of evidence of
resources by hypothesis means whether the
analyzing data tests used in (Unequal and different diet
distributions the world of unknown std plans differ
and performing Data Science dev) significantly
hypothesis and how to - Test of
testing choose the right independence
test for a given - One-way
business claim ANOVA
depending on
the associated
context.
10 2 Applied Graded Perform 4 Project - - - E-news Express Use A/B testing
Statistics statistical to identify the
analysis using effectiveness of
Python to the new
evaluate the landing page of
reliability of a an online news
particular portal
business
estimate using
confidence
intervals and
test hypotheses
(assumptions)
before putting
them into action
and committing
resources by
analyzing data
distributions
and performing
hypothesis
testing
11 3 Supervised Graded Understand the 1 Linear Machine 1. Intro to Learn the Used Car Price Explore and 1 8 7 4
Learning concept of Regression Learning (ML), a machine concept of Prediction visualize the
learning from subset of learning learning from data, build a
data, build Artificial 2. Simple Linear data, how the linear
linear and non- Intelligence Regression linear regression
linear models (AI), which 3. Multiple regression model to
to capture the focuses on Linear algorithm predict the
relationships developing Regression works, and how prices of used
between algorithms 4. Regression to build and cars, and
attributes and a capable of metrics assess the generate a set
known learning performance of of insights and
outcome, and patterns in data a linear recommendatio
evaluate the and making regression ns that will
performance of predictions model in help the
these models without being Python business
explicitly
programmed to
do so. Linear
Regression is
one of the most
popular
supervised ML
algorithms that
identifies the
degree of linear
relationship in
data. This
modules
introduces
participants to
ML and
explores how
linear
regression can
be used for
predictive
analysis.
12 3 Supervised Graded Understand the 2 Logistic Logistic - Introduction Understand the Credit Risk Analyze the
Learning concept of Regression regression is a to Logistic foundations of Analysis demographic
learning from statistical Regression the Logistic and socio-
data, build modeling - Interpretation Regression economic data
linear and non- technique from a Logistic model, how to of customers of
linear models primarily used Regression make a bank, build a
to capture the for modeling model interpretations predictive
relationships the probability - Changing the from the model. model to
between of binary threshold of a how to identify the
attributes and a outcomes and it Logistic evaluate the credibility of a
known finds Regression performance of loan application
outcome, and applications in model classification from a
evaluate the various fields - Evaluation of models, and customer, and
performance of such as a classification how changing help generate
these models medicine, model the threshold of insights to
finance, and - Pros and Cons a Logistic minimize loss
manufacturing. Regression from the bank’s
This module model can help perspective
covers the in improving
theory behind predictions
the logistic
regression
model, how to
asses its
performance,
and how to
draw statistical
inferences from
it.
13 3 Supervised Graded Understand the 3 Naive-Bayes, Bayes' Rule is - Bayes Rule Understand the Machine Analyze the
Learning concept of KNN an important - Naive Bayes principles and Predictive data of an auto
learning from topic in Algorithm applications of Maintenance component
data, build probabilistic - Distance Bayes' Rule, manufacturing
linear and non- reasoning and Metrics explore the company and
linear models decision- - KNN Naive Bayes develop a
to capture the making. Algorithm algorithm and predictive
relationships Distance its assumptions model to detect
between metrics offer a to apply it in potential
attributes and a handy way of classification machine
known measuring tasks, and gain failures,
outcome, and similarity proficiency in determine the
evaluate the between data implementing most
performance of points. This the K-Nearest influencing
these models module Neighbors factors on
provides (KNN) machine health,
participants algorithm for and provide
with a classification recommendatio
comprehensive and regression ns for cost
understanding tasks optimization to
of the Bayes the
Rule and Naive management
Bayes
algorithm, its
assumptions,
different
distance
metrics, the K-
Nearest
Neighbors
(KNN)
algorithm, and
its practical
applications in
classification
and regression
tasks.
14 3 Supervised Graded Understand the 4 SVM Support Vector - Support Understand the Income group DeltaSquare is
Learning concept of Machines Vector principles and classification an NGO that
learning from (SVMs) are a Machines mechanisms of works with the
data, build widely used (SVMs) Support Vector Government on
linear and non- supervised - Hard vs Soft Machines matters of
linear models learning Margin (SVMs), social policy to
to capture the algorithm for - Kernels in differentiate bring about a
relationships classification SVM between hard change in the
between and regression and soft margin lives of
attributes and a tasks. This SVMs and underprivileged
known module comprehend sections of
outcome, and provides their respective society. They
evaluate the participants advantages and are tasked with
performance of with a trade-offs, coming up with
these models comprehensive explore the a policy
understanding concept of framework by
of the concepts kernels in looking at the
of hard and soft SVMs, and gain data
margins in practical government got
SVMs and the experience in from WHO. The
importance of implementing objective is to
kernels like SVMs for analyze the
polynomial and inform data provided
RBF in SVMs to decision- to identify the
capture making and different factors
nonlinear problem- that influence
relationships in solving in real- the income of
data and world an individual, to
enhance model applications build a good
performance. predictive
model for
income and
assess it's
performance,
and help in
sharing a
proposal for the
government.
15 3 Supervised Graded Understand the 5 Project - - - Hotel Booking Analyze the
Learning concept of Cancellation booking data of
learning from Prediction a chain of
data, build hotels and
linear and non- build an ML
linear models model that will
to capture the help them
relationships predict the
between likelihood of a
attributes and a booking getting
known cancelled so
outcome, and that they can
evaluate the take measures
performance of to fill in
these models potential
vacancies and
reduce revenue
loss
16 - Break Week Self-paced - - AI Application AI is used 1. Application Experience how COVID Build an AI - - - -
Case Study across a variety of AI businesses use Detection solution for a
of businesses 2. AI Solution AI with tools & renowned chain
to enhance Workflow technologies of hospitals
operational 3. Measuring involved to that will help
efficiency, the impact of an solve problems them predict
streamline AI solution and gain an the likelihood of
decision- overview of a patient being
making how an end to infected by
processes, gain end AI solution COVID by
actionable works analyzing a
insights from chest X-ray
data, and foster scan of the
innovation, This patient to
module will segregate the
cover an end- patients who
to-end case are less likely
study to to have COVID
illustrate the and prioritize
different stages critical cases.
of
implementing
an AI solution
for a business
problem.
17 4 Ensemble Graded Combine the 1 Decision Trees Decision trees 1. Decision tree Learn the Wine Quality Analyze the 1 4 4 0
Techniques decisions from are supervised 2. Classification Decision Tree Prediction data of the
multiple ML algorithms metrics algorithm, how physiological
models using that utilize a 3. Pruning and a Decision Tree properties of
ensemble hierarchical Impurity is built, the different wine
techniques to structure for Measures different samples, build
improve model decision making 4. Regression pruning a predictive
performance and can be Trees techniques that model to
and make used for both 5. Logistic can be used to predict the
better classification Regression improve its quality of the
predictions, and and regression performance, wine using
employ feature problems. This about the these
engineering module dives different properties, and
techniques and into how a impurity identify the key
hyperparameter decison tree measures used, factors driving
tuning to arrive can be used to and how to the quality of
at generalized, model complex, identify the wine
robust models non-linear data most impactful
to optimize and how to attributes in
associated improve the decision-
business costs performance of making
decision trees
using pruning
techniques.
18 4 Ensemble Graded Combine the 2 Bagging and Random forest 1. Introduction Understand Diabetes Risk Build a
Techniques decisions from Random Forest is a popular to Ensemble how ensemble Prediction predictive
multiple ensemble Techniques techniques model to help
models using learning 2. Bagging work, the to identify
ensemble technique that 3. Random difference whether a
techniques to comprises of Forests between person is at risk
improve model several decision 4. Introduction bagging and of diabetes or
performance trees, each to Boosting boosting, and not
and make using a subset 5. Bagging VS build Random
better of the data to Boosting Forest and
predictions, and understand 6. Different Boosting
employ feature patterns. The boosting models to make
engineering outputs of each techniques - better
techniques and tree are then AdaBoost, predictions
hyperparameter aggregated to Gradient
tuning to arrive provide Boosting,
at generalized, predictive XGBoost
robust models performance.
to optimize Boosting
associated models are
business costs robust
ensemble
models that
comprise of
several sub-
models, each of
which are
developed in a
sequential
manner to
improve upon
the errors made
by the previous
one. This
modules will
covers random
forests and
essential
boosting
algorithms like
Adaboost and
XGBoost that
are widely used
in the industry
for accurate and
robust
predictions.
19 4 Ensemble Graded Combine the 3 Project - - - EasyVisa Facilitate the
Techniques decisions from process of visa
multiple approvals and
models using recommend
ensemble suitable profiles
techniques to for visa
improve model certification and
performance denial using
and make ensemble ML
better models
predictions, and
employ feature
engineering
techniques and
hyperparameter
tuning to arrive
at generalized,
robust models
to optimize
associated
business costs
20 5 Unsupervised Graded Understand the 3 K-mean K-means 1. Introduction Learn about the Bank Customer Identify 1 4 4 0
Learning concept of Clustering clustering is a to Clustering different types Segmentation different
learning from popular 2. K-Means of clustering segments in the
data, build unsupervised Clustering algorithms, existing
linear and non- ML algorithm 3. Hierarchical how K-means customer,
linear models that is used for Clustering clustering based on their
to capture the identifying 4. works, how to spending
relationships patterns in Dimensionality determine the patterns as
between unlabeled data Reduction optimal number well as past
attributes and a and grouping it. 5. PCA of clusters by interaction with
known This module comparing the bank, using
outcome, and dive into the different clustering
discover working of the metrics, and the algorithms and
patterns and algorithm and importance of provide
segment data the important scaling data recommendatio
with no labels points to keep ns to the bank
in mind when on how to
implementing it better market
in practical to and service
scenarios. these
customers.
21 5 Unsupervised Graded Explore 4 Hierarchical Hierarchical - Hierarchical Learn how to Tourism Tourism is now
Learning different Clustering and clustering Clustering apply the Services recognized as a
supervised PCA organizes data - Cophenetic hierarchical directly
learning into a tree-like Correlation clustering measurable
algorithms like structure of - Introduction technique to activity,
logistic nested clusters, to group similar enabling more
regression and while Dimensionality data points accurate
support vector dimensionality Reduction together and analysis and
machines reduction - Principal discover more effective
(SVMs), and techniques are Component underlying policies can be
unsupervised used to Analysis patterns, made for
learning transform data understand the tourism.
algorithms like into a lower- need for Whereas
hierarchical dimensional reducing previously the
clustering) and space while dimensions of sector relied
principal retaining the the data, and mostly on
component most important understand the approximations
analysis (PCA), information in working of the from related
and apply and it. This module PCA and how areas of
evaluate the covers the to transform measurement
machine business data into fewer (e.g. Balance of
learning applications of dimensions Payments
techniques and hierarchical using PCA statistics),
algorithms to clustering and tourism
real-world how to reduce nowadays is a
datasets and the dimension productive
solve business of data using activity that can
problems PCA to aid in be analyzed
visualization using factors
and feature like economic
selection of indicators,
multivariate social
datasets. indicators,
environmental
& infrastructure
indicators, etc.
The task is to
analyze several
of these factors
and group
countries based
on them to help
understand the
key locations
where the
company can
invest to
promote
tourism
services.
22 5 Unsupervised Graded Combine the 4 Project - - - Trade&Ahead Build a
Learning decisions from diversified
multiple portfolio by
models using analyzing and
ensemble clustering
techniques to stocks based on
improve model financial
performance attributes
and make
better
predictions, and
employ feature
engineering
techniques and
hyperparameter
tuning to arrive
at generalized,
robust models
to optimize
associated
business costs
23 6 Featurization, Graded Employ feature 1 Feature Feature - Feature Learn how to Job change An Ed-Tech 1 4 7 0
Model Selection engineering Engineering engineering Engineering handle prediction company wants
& Tuning techniques and and Cross involves - Cross- imbalanced to hire data
hyperparameter Validation creating new validation data, how to scientists
tuning to input features - Oversampling use the cross- among people
improve model or modifying and validation who have
performance existing ones to Undersampling technique to successfully
and optimize improve a - Regularization get a better passed some
associated machine picture of courses and
business costs learning model then signed up
model's performance, for training. The
performance, and understand company wants
and cross- the concept of to know which
validation is regularization of the people
used for getting are really
a better looking for a job
assessment of a change and will
model prefer working
performance. with them, after
This module completion of
covers these training,
two concepts because it
along with helps to reduce
regularization the cost and
to tune the time for
performance of categorization
ML models and of candidates.
correctly assess Information
their related to
performance. demographics,
education,
experience is in
hands from
candidates
signup and
enrollment. The
objective is to
identify the
factors
affecting a
person looking
for a job
change, build a
predictive
model to
predict whether
a person is
looking for a job
change, and
check whether
imbalance in
the data affects
model
predictions.
24 6 Featurization, Graded Employ feature 2 ML Pipeline and Hyperparamete - Machine Learn how to Supermarket All You Need'
Model Selection engineering Hyperparamete r tuning Learning optimize model marketing Supermarket is
& Tuning techniques and r Tuning involves Pipeline performance campaign planning for the
hyperparameter optimizing the - Model Tuning using year-end sale -
tuning to configuration of and hyperparameter they want to
improve model a machine Performance tuning and how launch a new
performance learning model - to automate offer i.e. gold
and optimize to enhance its Hyperparamete standard membership for
associated performance. r Tuning workflows in a only $499 that
business costs This module - Grid Search machine is of $999 on
covers two - Random learning normal days
common Search process using (that gives 20%
techniques to pipelines discount on all
find the optimal purchases) only
hyperparameter for existing
s of an ML customers, for
model given a that they need
business to do a
context, and campaign
how to create through phone
an ML pipeline calls - the best
to conduct data way to reduce
processing and the cost of the
modeling in a campaign is to
streamlined make a
and predictive
reproducible model to
manner. classify
customers who
might purchase
the offer, using
the data they
gathered during
last year's
campaign.
The objective is
to build a
model for
classifying
whether
customers will
respond
positively or
not, identfiy the
different factors
which affect the
kind of
response, and
to improve the
performance of
an initially built
model using
hyperparameter
tuning.
25 6 Featurization, Graded Employ feature 3 Project - - - Credit Card Analyze the
Model Selection engineering Users Churn data and come
& Tuning techniques and Prediction up with a
hyperparameter predictive
tuning to model to
improve model determine if a
performance customer will
and optimize leave the credit
associated card services or
business costs not and the
reason behind it
26 7 Introduction to Graded Understand the 1 Querying data SQL is a widely - Getting set up Understand the GL Beats Query the data 1 3 3 4
SQL core concepts with SQL used querying with MySQL core concepts from the
of databases language for - Introduction of databases database of a
and SQL, gain efficiently to DB and SQL and SQL, music-based
practical managing and - Fetching data including startup that
experience manipulating in SQL tables, rows, recently started
writing simple relational - Filtering data columns, and selling music
SQL queries to databases. This in SQL SQL records, answer
filter, module commands, questions for a
manipulate, and provides an apply SQL performance
retrieve data essential queries to review to
from relational foundation for retrieve data identify
databases, and understanding from tables, customer
utilize complex and working and utilize preferences by
SQL queries with relational filtering demographies,
with joins, databases. techniques in and generate
window Participants SQL to extract recommendatio
functions, and will explore the specific data ns to help
subqueries for fundamentals subsets based business
dynamic data of setting up on defined growth
extraction and MySQL, criteria,
manipulation to including enhancing data
solve real- installation and analysis and
world data configuration, decision-
problems and gain insight into making
extract the principles of capabilities
actionable database
business management
insights and Structured
Query
Language
(SQL), and
learn how to
fetch and filter
data using SQL
queries,
enabling them
to extract
valuable
insights from
large datasets
efficiently.
27 7 Introduction to Graded Understand the 2 SQL In-Built SQL offers a - Numeric Demonstrate GL Beats Query the data
SQL core concepts Functions wide range of Functions in proficiency in from the
of databases numeric, string, SQL utilizing database of a
and SQL, gain and date - String numeric, string, music-based
practical functions, Functions in and date startup that
experience gaining SQL functions to recently started
writing simple proficiency in - Date perform selling music
SQL queries to leveraging Functions in mathematical records, answer
filter, these functions SQL operations and questions for a
manipulate, and to perform - Aggregating calculations, performance
retrieve data advanced data in SQL manipulate and review to
from relational calculations, format text identify
databases, and string data, and customer
utilize complex manipulations, perform date preferences by
SQL queries and date arithmetic, and demographies,
with joins, operations. This gain an and generate
window module understanding recommendatio
functions, and provides a of aggregating ns to help
subqueries for comprehensive data using SQL business
dynamic data exploration of functions to growth
extraction and the various summarize and
manipulation to functions analyze
solve real- available within datasets,
world data SQL for data enabling
problems and manipulation informed
extract and analysis. decision-
actionable Additionally, making and
business participants will data-driven
insights discover the insights.
significance of
aggregating
data using SQL
functions,
enabling them
to summarize
and analyze
large datasets
effectively.
28 7 Introduction to Graded Understand the 3 Advanced SQL joins are - Joins in SQL Understand the HR Data Query job
SQL core concepts Querying used to - Window purpose and Analysis history data of
of databases combine data functions in mechanics of an organization
and SQL, gain from multiple SQL joins in SQL, to identify
practical tables - Subqueries apply window employees'
experience effectively and functions to tenure, analyze
writing simple window perform salary
SQL queries to functions advanced differentials
filter, enable analytical based on job
manipulate, and performing operations such titles, and
retrieve data complex as ranking and evaluate
from relational analytical tasks partitioning departmental
databases, and such as ranking, data within workforce
utilize complex partitioning, specified distribution to
SQL queries and windows, and gain practical
with joins, aggregating utilize insights for
window data within subqueries to strategic
functions, and specified nest queries decision-
subqueries for windows. within other making and
dynamic data Subqueries queries to workforce
extraction and allow one to enable dynamic optimization
manipulation to nest queries data extraction
solve real- within other and
world data queries. This manipulation to
problems and module will solve real-
extract equip world data
actionable participants analysis
business with advanced challenges and
insights techniques for extract
querying and actionable
analyzing insights from
relational relational
databases to databases.
extract and
manipulate
data
dynamically.
29 7 Introduction to Graded Understand the 4 Project - - - New Wheels Analyze the
SQL core concepts Data Analysis listing and
of databases customer
and SQL, gain feedback data
practical of a vehicle
experience resale
writing simple company,
SQL queries to answer an array
filter, of questions,
manipulate, and and create a
retrieve data quarterly
from relational business report
databases, and for the
utilize complex leadership to
SQL queries enable data-
with joins, driven decision
window making
functions, and
subqueries for
dynamic data
extraction and
manipulation to
solve real-
world data
problems and
extract
actionable
business
insights
30 - Break Week Self-paced - - AI Application AI is used 1. Application Experience how Airline Analyze the - - - -
Case Study across a variety of AI businesses use Customer reviews in
of businesses 2. AI Solution AI with tools & Review tweets shared
to enhance Workflow technologies Sentiment by customers of
operational 3. Measuring involved to Analysis an airline
efficiency, the impact of an solve problems regarding their
streamline AI solution and gain an experience with
decision- overview of the services
making how an end to provided and
processes, gain end AI solution build a
actionable works predictive
insights from model that can
data, and foster parse the
innovation, This customer
module will reviews and
cover an end- predict the
to-end case sentiment of
study to the customer
illustrate the
different stages
of
implementing
an AI solution
for a business
problem.
31 8 Neural Graded Implement 1 Introduction to Artificial Neural 1. Overview of Demonstrate an University Analyze the 1 5 5 3
Networks neural Neural Networks Neural understanding Admission data and build
networks to Networks (ANN) are Networks of the working Likelihood a deep neural
synthesize inspired by 2. Multi Layer of an artificial Prediction network using
knowledge biological Perceptron neural network, different
from data, neurons and 3. Types of design and attributes listed
demonstrate an utilize Activation implement on candidate
understanding collection of Functions neural profiles to
of different artificial 4. Training the networks to predict the
optimization neurons Neural Network synthesize chances of
algorithms and stacked and knowledge getting
regularization connected in from data, and admitted to the
techniques, and layers to model evaluate the Masters
evaluate the complex data. factors that program of a
factors that This module contribute to university
contribute to dives deeper improving the
improving into the performance of
performance to underlying neural
build functionality of networks to
generalized and neural solve business
robust neural networks and problems
network how to use
models to solve common, open-
business source Deep
problems Learning (DL)
libraries Keras
and Tensorflow
to build neural
networks to
solve business
problems.
32 8 Neural Graded Implement 2 Optimizing Different 1. Optimizers Demonstrate an Credit Card Analyze credit
Networks neural Neural optimization 2. Weight understanding Fraud Detection card transaction
networks to Networks algorithms are Initialization of different data and build
synthesize used in 3. optimization a neural
knowledge practical setups Regularization algorithms and network model
from data, to improve the 4. Different regularization to capture the
demonstrate an performance of Neural Network techniques, and complexities in
understanding neural Models apply the same the data and
of different networks, and to build predict the
optimization deploying generalized and probability of a
algorithms and regularization robust neural transaction
regularization techniques like network being
techniques, and dropout and models to solve fraudulent to
evaluate the batch business help minimize
factors that normalization problems financial losses
contribute to helps in incurred by the
improving the ensuring neural financial
performance to networks gain a institution and
build generalize better the cardholders
and robust understanding
neural network of the patterns
models to solve in complex
business data. This
problems module dives
deeper into the
concepts of
optimization
and
regularization
to illustrate the
best practices
for building
generalized and
robust neural
networks to
solve business
problems.
33 8 Neural Graded Implement 3 Project - - - Signal Quality Analyze the
Networks neural Prediction data provided
networks to by a
synthesize communication
knowledge s equipment
from data, manufacturing
demonstrate an regarding a
understanding product which
of different is responsible
optimization for emitting
algorithms and informative
regularization signals, and
techniques, and build a deep
evaluate the learning model
factors that which can help
contribute to the company to
improving the predict the
performance to equipment’s
build generalize signal quality
and robust using various
neural network parameters
models to solve
business
problems
34 9 Computer Graded Get introduced 1 Introduction to Computer 1. Overview of Get introduced Image Apply the 1 5 7 0
Vision to the world of Image Vision is a Computer to the world of Processing different image
computer Processing branch of AI Vision image processing like
vision, that focuses on 2. Color Pixel processing, color channel
demonstrate an understanding Theory and demonstrate an manipulation,
understanding and extracting Image understanding different filters,
of image meaningful Representation of the edge detection,
processing and insights from 3. Edge fundamental padding, and
different image data. Detection and theories behind more to extract
methods to This module Kernels image informatice
extract provides an 4. Padding, representation features from
informative overview of the Strides, and and image an image
features from world of Pooling processing, and
images, build computer vision gain a practical
convolutional and covers understanding
neural techniques to of different
networks process images ways to extract
(CNNs) to and extract informative
unearth hidden meaningful features from
patterns in patterns from images
image data, and them.
leverage
common CNN
architectures to
solve image
classification
problems
35 9 Computer Graded Get introduced 2 Introduction to Given the 1. ANNs vs Analyze the Food Image Build an image
Vision to the world of Convolutional complex nature CNNs differences Classification classifier using
computer Neural of image data, 2. Introduction between CNNs to tag the
vision, Networks convolutional to CNN Artificial Neural food-related
demonstrate an neural Architecture Networks images of a
understanding networks (ANNs) and stock
of image (CNNs) are Convolutional photography
processing and utilized to Neural company for
different capture Networks the ease of
methods to relevant spatial (CNNs) in the access of
extract information in context of customers
informative images. This computer vision
features from module will tasks, design
images, build cover the and implement
convolutional fundamentals basic CNN
neural of CNNs and models for
networks how to build specific
(CNNs) to them from computer vision
unearth hidden scratch to build tasks, and
patterns in powerful evaluate the
image data, and models for factors that
leverage image contribute to
common CNN classification. improving the
architectures to performance
solve image
classification
problems
36 9 Computer Graded Get introduced 3 Transfer Transfer 1. Demonstrate an Monkey Species Build an image
Vision to the world of Learning learning is a Regularization understanding Classification classifier using
computer method to in CNN's of pre-trained
vision, leverage the 2. Introduction regularization CNN
demonstrate an underlying to Transfer and transfer architectures
understanding knowledge Learning learning in the for tagging
of image needed to solve 3. Common context of different
processing and one problem to CNN CNNs, apply species of
different other problems. Architectures transfer monkeys based
methods to This module learning to on collected
extract covers the adapt pre- image data to
informative concept of trained models help a wildlife
features from transfer to specific tasks sanctuary in
images, build learning in the by fine-tuning, identifying and
convolutional context of and assess the gathering data
neural computer vision suitability of on currently
networks and how to transfer endangered
(CNNs) to leverage learning for species
unearth hidden common CNN specific
patterns in architectures business
image data, and like VGG16 and scenarios
leverage InceptionNet to
common CNN solve different
architectures to image
solve image classification
classification problems.
problems
37 9 Computer Graded Understand the 4 Object Object - Introduction Understand the Autonomous Build a
Vision fundamentals Detection detection to Object concept and Car Scene computer vision
of object involves the Detection importance of Understanding system
detection and detection of - Different object detection powered by
segmentation objects within algorithms for in computer neural
and their images or Object vision tasks. networks for a
importance in videos, and Detection Recognize car
computer vision plays a - Applications common manufacturer
applications, significant role of Object challenges and that will enable
compare and in various Detection considerations a self-driving
contrast applications. associated with car to sense its
different This module object surroundings,
algorithms and provides detection, identify objects
approaches for participants including in its
object detection with a varying object neighborhood,
and comprehensive sizes, and act safely
segmentation, overview of occlusions, and to reach a
and apply object detection background certain target
object detection algorithms and clutter.
and methodologies Identify and
segmentation like R-CNNs describe
to solve real- and YOLO, their different
world computer strengths and algorithms and
vision problems weaknesses, techniques
and real-world used for object
applications. detection, such
as region-
based methods,
single-shot
detectors
(SSDs), and
two-stage
detectors (e.g.,
Faster R-CNN,
YOLO).
Explore real-
world
applications of
object detection
across domains
such as
autonomous
vehicles,
surveillance
systems,
medical
imaging, and
retail analytics.
Gain insights
into the
practical
implementation
and
deployment of
object detection
systems,
including data
collection,
model training,
evaluation
metrics, and
performance
optimization.
Apply
knowledge of
object detection
algorithms and
applications to
address specific
use cases and
challenges
through hands-
on projects and
38 9 Computer Graded Understand the 5 Object Image - Introduction Understand the Autonomous Build a
Vision fundamentals Segmentation segmentation to Image fundamental Car Scene computer vision
of object plays a pivotal Segmentation concepts of Understanding system
detection and role in - Semantic image powered by
segmentation computer vision Segmentation segmentation neural
and their tasks such as - Instance and its role in networks for a
importance in object Segmentation computer vision car
computer vision localization and - Siamese tasks. manufacturer
applications, scene Networks Differentiate that will enable
compare and understanding. - Applications between a self-driving
contrast This module of Image various types of car to sense its
different provides Segmentation image surroundings,
algorithms and participants segmentation conduct scene
approaches for with a techniques, understanding,
object detection comprehensive including and act safely
and overview of semantic to reach a
segmentation, different types segmentation, certain target
and apply of image instance
object detection segmentation, segmentation,
and including and their
segmentation semantic respective
to solve real- segmentation methodologies.
world computer and instance Explore
vision problems segmentation, advanced topics
along with such as
advanced Siamese
concepts such networks and
as Siamese their
networks. applications in
Participants tasks such as
will gain image similarity
insights into the and one-shot
diverse learning.
applications of Recognize the
image significance of
segmentation image
with the help of segmentation in
real-world real-world
examples and applications,
case studies. including
medical image
analysis,
autonomous
driving, satellite
imagery, and
industrial
automation.
Gain practical
experience in
implementing
and deploying
image
segmentation
models for
specific use
cases and
challenges
through hands-
on projects and
exercises.
Analyze and
evaluate the
performance of
image
segmentation
algorithms
using
appropriate
metrics and
techniques,
ensuring their
suitability for
specific tasks
39 9 Computer Graded Understand the 6 Project - - - Face Detection Analyze the
Vision fundamentals and data of a
of object Recognition company that
detection and owns a movie
segmentation application and
and their repository that
importance in caters movie
computer vision streaming to
applications, millions of
compare and users
contrast worldwide and
different build a
algorithms and computer vision
approaches for system
object detection powered by
and neural
segmentation, networks to
and apply automate the
object detection process of cast
and and crew
segmentation identification
to solve real- for different
world computer scenes in
vision problems movies so that
relevant
information can
be shared
about them
40 10 Natural Graded Get introduced 1 Introduction to Natural 1. Introduction Demonstrate an Airline Analyze the 1 4 6 0
Language to the world of Natural Language to NLP understanding Customer reviews in
Processing natural Language Processing 2. History of of the intuition Review tweets shared
language Processing (NLP) is a NLP behind NLP by Sentiment by customers of
processing, branch of AI 3. Applications applying basic Analysis an airline
gain a practical that focuses on of NLP principles to regarding their
understanding processing and 4. Text explain how experience with
of text understanding Processing language the services
processing and human 5. Text processing provided and
vectorization language to Vectorization tasks are build a
methods, gain a facilitate the 6. Sentiment approached, predictive
practical interaction of Analysis gain a practical model that can
understanding machines with understanding parse the
of the working it. This module of the need for customer
of different introduces sentiment reviews and
transformer learners to the analysis, and predict the
architectures world of NLP apply text sentiment of
that lie at the and covers processing and the customer
core of large different text vectorization
language processing and methods to
models (LLMs), vectorization solve business
and design and techniques to problems
implement efficiently
robust NLP extract
solutions using information
open-source from raw
LLMs combined textual data.
with prompt
engineering
techniques
41 10 Natural Graded Get introduced 2 Word Word 1. Introduction Evaluate the Product Analyze the
Language to the world of Embeddings embeddings to Word advantages and Reviews data of reviews
Processing natural allow us to Embeddings limitations of Sentiment shared by
language numerically 2. Word2Vec different word Analysis customers
processing, represent 3. GloVe embedding regarding their
gain a practical complex textual 4. Semantic methods and experience with
understanding data, thereby Search design and the products of
of text enabling us to implement a rising e-
processing and perform a hands-on commerce
vectorization variety of solutions platform and
methods, gain a operations on combining develop an AI
practical them. This word solution that
understanding module will embedding can
of the working cover different methods with automatically
of different word machine assess and
transformer embedding learning categorize the
architectures techniques and techniques for sentiment of
that lie at the the steps solving NLP the customers.
core of large involved in problems
language designing and
models (LLMs), implementing
and design and hands-on
implement solutions
robust NLP combining
solutions using word
open-source embedding
LLMs combined methods with
with prompt machine
engineering learning
techniques techniques for
solving NLP
problems
42 10 Natural Graded Get introduced 3 Attention Transformers 1. Introduction Demonstrate a E-news Efficiently
Language to the world of Mechanism and are neural to Transformers working Platform News categorize and
Processing natural Transformer network 2. Components knowledge of Categorization tag news
language Models architectures of a the architecture articles for an
processing, that develop a Transformer of an Encoder- e-news
gain a practical context-aware 3. Different Decoder platform for
understanding understanding Transformer Transformer by improved
of text of data and Architectures breaking down content
processing and have 4. Applications the roles of the organization
vectorization revolutionized of Transformers encoder and and user
methods, gain a the field of NLP decoder engagement
practical by exhibiting components,
understanding exceptional evaluate the
of the working performance performance
of different across a wide and suitability
transformer variety of tasks. of various
architectures This module transformer
that lie at the dives into the architectures
core of large underlying for specific
language workings of tasks, and
models (LLMs), different design and
and design and transformer implement NLP
implement architectures solutions to
robust NLP and how to use business
solutions using them to solve problems using
open-source complex NLP different
LLMs combined tasks. transformer
with prompt architectures
engineering
techniques
43 10 Natural Graded Get introduced 4 Large Large 1. Introduction Demonstrate an Restaurant Analyze the
Language to the world of Language Language to LLMs understanding Review provided
Processing natural Models and Models (LLMs) 2. Working of of the Analysis sample of the
language Prompt are ML models LLMs capabilities and customer
processing, Engineering that are pre- 3. Applications limitations of reviews for
gain a practical trained on large of LLMs Large different
understanding corpora of data 4. Introduction Language restaurants for
of text and possess to Prompt Models (LLMs), a leading global
processing and the ability to Engineering design, food
vectorization generate 5. Strategies for implement, and aggregator and
methods, gain a coherent and devising evaluate build a
practical contextually prompts tailored predictive
understanding relevant prompts for model to
of the working content. Prompt specific tasks analyze the
of different engineering is a for achieving reviews and tag
transformer process of desired them, thereby
architectures iteratively outcomes, and enhancing the
that lie at the deriving a solve NLP company's
core of large specific set of problems by ability to
language instructions to integrating the understand
models (LLMs), help an LLM understanding customer
and design and accomplish a of open-source sentiments at
implement specific task. LLM scale, enabling
robust NLP This module capabilities and data-driven
solutions using introduces prompt decision-
open-source LLMs, explains engineering making, and
LLMs combined their working, improving
with prompt and covers overall
engineering practices to customer
techniques effectively satisfaction.
devise prompts
to solve
problems using
LLMs.
44 10 Natural Graded Get introduced 5 Project - - - Customer Explore the
Language to the world of Support Ticket data provided
Processing natural Categorization and develop an
language advanced
processing, support ticket
gain a practical categorization
understanding system that
of text accurately
processing and classifies
vectorization incoming
methods, gain a tickets, assigns
practical relevant tags
understanding based on their
of the working content, and
of different assigns priority
transformer and ETA for the
architectures tickets for
that lie at the prompt
core of large resolution
language
models (LLMs),
and design and
implement
robust NLP
solutions using
open-source
LLMs combined
with prompt
engineering
techniques
45 - Break Week Self-paced - - - - - - - - - - - -
46 11 Capstone Graded Identify and 1 - - - - - - 1 0 0 0
define a real-
world problem
considering
factors such as
data
availability,
feasibility, and
potential
impact, design
and develop an
AI solution that
addresses the
identified
problem,
explore,
analyze, and
process the
data, apply and
evaluate
appropriate AI
techniques to
implement the
solution
effectively, and
communicate
insights and
implications to
stakeholders
47 11 Capstone Graded Identify and 2 - - - - - -
define a real-
world problem
considering
factors such as
data
availability,
feasibility, and
potential
impact, design
and develop an
AI solution that
addresses the
identified
problem,
explore,
analyze, and
process the
data, apply and
evaluate
appropriate AI
techniques to
implement the
solution
effectively, and
communicate
insights and
implications to
stakeholders
48 11 Capstone Graded Identify and 3 Interim - - - - -
define a real- Submission
world problem
considering
factors such as
data
availability,
feasibility, and
potential
impact, design
and develop an
AI solution that
addresses the
identified
problem,
explore,
analyze, and
process the
data, apply and
evaluate
appropriate AI
techniques to
implement the
solution
effectively, and
communicate
insights and
implications to
stakeholders
49 11 Capstone Graded Identify and 4 - - - - - -
define a real-
world problem
considering
factors such as
data
availability,
feasibility, and
potential
impact, design
and develop an
AI solution that
addresses the
identified
problem,
explore,
analyze, and
process the
data, apply and
evaluate
appropriate AI
techniques to
implement the
solution
effectively, and
communicate
insights and
implications to
stakeholders
50 11 Capstone Graded Identify and 5 Final - - - - -
define a real- Submission
world problem
considering
factors such as
data
availability,
feasibility, and
potential
impact, design
and develop an
AI solution that
addresses the
identified
problem,
explore,
analyze, and
process the
data, apply and
evaluate
appropriate AI
techniques to
implement the
solution
effectively, and
communicate
insights and
implications to
stakeholders
51 11 Capstone Graded Identify and 6 Final - - - - -
define a real- Presentation
world problem
considering
factors such as
data
availability,
feasibility, and
potential
impact, design
and develop an
AI solution that
addresses the
identified
problem,
explore,
analyze, and
process the
data, apply and
evaluate
appropriate AI
techniques to
implement the
solution
effectively, and
communicate
insights and
implications to
stakeholders
- - Recommendati Self-Paced Get introduced - - - 1. Intro to - E-commerce Build a 1 2 2 2
on System to Recommendati Product recommendatio
recommendatio on Systems Recommendati n system for an
n systems and 2. Market on e-commerce
learn how to Basket Analysis website to
build 3. Popularity provide
recommendatio Based and personalized
n systems that Content Based product
use past Recommendati recommendatio
product on Systems ns for the
purchase 4. Collaborative customers
and satisfaction Filtering
data to make 5. Hybrid
high-quality Recommendati
personalized on Systems
recommendatio
ns
- - Model Self-Paced Gain an - - - 1. Introduction - - - 0 0 0 0
Deployment overview of to Model
model deployment
deployment, 2. Model
how to serialize Serialization -
ML models to Pickling
make them 3. Batch Mode
accessible, and and Flask
use 4. Docker and
technologies Kubernetes
like Docker and
Kubernetes for
scaling
deployed ML
models
- - GANs Self-paced Understand the - - - - Introduction - - - 0 0 0 0
fundamental to GANs
concepts and - Working of
principles of GANs
Generative - KL & JS
Adversarial Divergence
Networks - Types of
(GANs), gain a GANs
conceptual - Evaluating
understanding GANs
of how they - Applications
generate of GANs
realistic
synthetic data,
gain proficiency
in using
information
theory concepts
such as
Kullback-
Leibler (KL)
divergence and
Jensen-
Shannon (JS)
divergence to
guide the
training of
GANs,
understand
their
architecture of
different types
of GANs, and
apply GANs to
real-world
problem
scenarios
- - Time Series Self-paced Learn how to - - - - Introduction - - - 0 0 4 3
Forecasting describe to Time Series
components of - Forecasting
a time series models
data and - Exponential
analyze them Smoothing
using special - Stationarity
techniques and - MA, ARMA,
methods for ARIMA,
time series SARIMA
forecasting
- - Reinforcement Self-paced Understand the - - - - Introduction - - - 0 0 0 0
Learning fundamental to
principles and Reinforcement
components of Learning
reinforcement - Reinforcement
learning (RL), Learning
explore the Framework
working - Working of
mechanisms of the different
different components of
components of Reinforcement
RL, gain - Q-Learning
proficiency in - SARSA
implementing Algorithm
Q-learning and
SARSA
algorithms, and
apply the
algorithms to
real-world
reinforcement
learning tasks
- - Visualization Self-paced Learn how to - - - - Tensorboard - - - 0 0 0 0
using utilize - Visualizing
TensorBoard TensorBoard to weights, bias,
visualize and and gradients
monitor deep - Occlusion
learning experiment
models by - Saliency maps
visualizing loss - Neural style
curves, transfer
performance
metrics,
weights, biases,
and gradients
of neural
networks,
conduct
occlusion
experiments to
assess model
interpretability,
generate
saliency maps
to visualize
important
features, and
briefly explore
neural style
transfer
techniques to
transform
images into
artistic styles
using neural
networks