SHREEYASH PRATISHTHAN’S

SHREEYASH COLLEGE OF ENGINEERING AND TECHNOLOGY

(POLYTECHNIC), CHH. SAMBHAJINAGAR

MICRO-PROJECT REPORT

NAME OF DEPARTMENT:- ARTIFICIAL INTELLIGENCE AND

MACHINE LEARNING
ACADEMIC YEAR:- 2024-25
SEMESTER:- 5TH
COURSE NAME :- CLOUD COMPUTING FOR DATA
SCIENCE

COURSE CODE:
MICRO-PROJECT TITLE:- CLOUD COMPUTING IN MACHINE
LEARNING
PREPARED BY:-

1) NIKITA SHANKAR HINGMIRE ENROLL.NO:-2210920387

UNDER THE GUIDANCE OF :- Prof. T.A.GAVHAL

MAHARASHTRA STATE BOARD OF TECHNICAL EDUCATION,
MUMBAI
CERTIFICATE

This is to certify that Mr./ Ms Nikita Shankar hingmire of 5th Semester of

Diploma in Artificial Intelligence and Machine Learning of Institute Shreeyash
College Of Engineering And Technology (polytechnic) has successfully completed
Micro-Project Work in Course of Cloud computing for data science for the
academic year 2024-25 as prescribed in the I-Scheme Curriculum.

Date:- Enrollment No:- 2210920387

Place:- chh.sambhajinagaar Exam Seat No.:-

signatur Signature Signature

e
Guide HOD Principal
 ACKNOWLEDGEMENT

We wish to express our profound gratitude to our guide

Prof. T.A.GAVHAL who guided us endlessly in framing and
completion of Micro-Project. He / She guided us on all the main points
in that Micro-Project. ssWe are indebted to his / her constant
encouragement, cooperation and help. It was his / her enthusiastic
support that helped us in overcoming of various obstacles in the Micro-
Project.
We are also thankful to our Principal, HOD, Faculty Members
and classmates for extending their support and motivation in the
completion of this Micro-Project.

1) NIKITA SHANKAR HINGMIRE ENROLL.NO:- 2210920387

 Micro-Project Proposal

Title of Micro-Project:- Cloud Computing in Machine Learning

Aims/Benefits of the Micro-Project :

1. Scalable Infrastructure: Provide scalable computing resources for machine learning
workloads.

2. Faster Processing: Accelerate machine learning processing using cloud-based

parallel processing.

3. Cost-Effectiveness: Reduce costs associated with machine learning infrastructure

and maintenance.

4. Collaboration: Enable collaboration among data scientists and researchers.

5. Accessibility: Make machine learning capabilities accessible to organizations of all

sizes.

Course Outcomes Addressed:

1. Understand cloud computing concepts and services

2. Familiarity with machine learning algorithms and models

3. Knowledge of cloud-based machine learning platforms and tools

4. Understanding of distributed computing and parallel processing

5. Familiarity with cloud security and compliance.

Proposed Methodology:
1. Problem Definition

- Identify business problem or research question

- Define machine learning goals and objectives

2. Data Collection

- Gather relevant data from various sources

- Preprocess data for machine learning algorithms

3. Cloud Platform Selection

- Choose suitable cloud provider (e.g., AWS, Azure, Google Cloud)

- Select appropriate cloud services (e.g., SageMaker, Azure ML)

4. Model Development

- Develop and train machine learning models

- Use cloud-based frameworks (e.g., TensorFlow, PyTorch)

5. Model Deployment

- Deploy models on cloud platforms

- Use containerization (e.g., Docker) and orchestration (e.g., Kubernetes)needed.

Resources Required

Sr. Name of Resources / Materials Specification Qty Remarks

No.
1 Computer RAM minimum 1
4GB,i5 7th Gen

2 Operating system Windows11 1

3 Internet Google 1

Names of Team Members with En. Nos.

1.Nikita Shankar Hingmire En.no 2210920387

(To be approved by the concerned teacher)

PROF.T.A.GAVHAL.

Action Plan :-
 Name of
Sr. Planned Planne Responsible
Details of activity
No. Week Start d Team Members
date Finish
date

1 1 &2 Discussion & Finalization of 02/09/24 05/09/24 Self

Topic

2 3 Preparation of the Abstract 05/09/24 06/09/24 Self

3 4 Literature Review 06/09/24 08/09/24 Self
4 5 Submission of Microproject 09/09/24 10/09/24 Self
Proposal ( Annexure-I)

5 6 Collection of information about 11/09/24 13/09/24 Self

Topic

6 7 Collection of relevant content / 14/09/24 16/09/24 Self

materials for the execution of
Microproject.

7 8 Discussion and submission of 17/09/24 19/09/24 Self

outline of the Microproject.

8 9 Analysis / execution of Collected 20/09/24 25/09/24 self

data / information and
preparation of Prototypes /
drawings / photos / charts /
graphs / tables / circuits / Models
/ programs etc.

9 10 Completion of Contents of 26/09/24 30/09/24 Self

Project Report

10 11 Completion of Weekly progress 01/10/24 05/10/24 Self

Report

11 12 Completion of Project Report ( 06/10/24 10/10/24 Self

Annexure-II)

12 13 Viva voce / Delivery of self

Presentation

Micro-Project Report
Title of Micro-Project:- Cloud computing in machine learning.

Rationale : Cloud computing enhances machine learning capabilities, making it

faster, scalable, and cost-effective.
By leveraging cloud computing, machine learning models can be trained faster,
deployed efficiently and scaled effortlessly, driving business innovation.
Cloud computing has revolutionized the field of machine learning by providing:

1. Scalability: Cloud computing resources can be scaled up or down to handle large

datasets and complex computations.

2. Flexibility: Cloud providers offer a range of machine learning frameworks, tools,

and services.

3. Cost-effectiveness: Cloud computing reduces infrastructure and maintenance costs.

4. Faster Processing: Cloud computing leverages parallel processing and distributed

computing.

5. Collaboration: Cloud computing enables seamless teamwork and data sharing.

Aims/Benefits of the Micro-Project:-

Aims:

1. Scalable Infrastructure
2. Faster Processing
3. Cost-Effectiveness
4. Collaboration
5. Accessibility

Benefits:

1. Scalability
2. Flexibility
3. Automation
4. Integration
5. Security

3.0 Course Outcomes Achieved

1. Understand cloud computing concepts and services

 2. Familiarity with machine learning algorithms and models
3. Knowledge of cloud-based machine learning platforms and tools
4. Understanding of distributed computing and parallel processing
5. Familiarity with cloud security and compliance.

5.0 Actual Methodology Followed

Phase 1: Data Preparation
1. Data Collection: Gathering data from various sources.
2. Data Preprocessing: Cleaning, transforming, and preparing data for
modeling.
- Concepts: Data Wrangling, Data Munging, Feature Engineering.
- Tools: Pandas, NumPy, Apache Spark.

Phase 2: Model Development

1. Model Selection: Choosing suitable machine learning algorithms.
2. Model Training: Training models using cloud-based resources.
- Concepts: Supervised/Unsupervised Learning, Deep Learning, Neural
Networks.
- Tools: TensorFlow, PyTorch, Scikit-Learn.

Phase 3: Model Deployment

1. Model Deployment: Deploying models using cloud-based services.
- Concepts: Containerization, Orchestration, Cloud Native
Applications.
- Tools: Docker, Kubernetes, Cloud Foundry.
2. Model Monitoring: Monitoring model performance and resource
utilization.
- Concepts: Logging, Monitoring, Alerting.
- Tools: Prometheus, Grafana, New Relic.

Phase 4: Model Maintenance

1. Model Updates: Updating models to adapt to changing data.
2. Model Optimization: Optimizing model performance and resource
utilization.
- Concepts: Hyperparameter Tuning, Model Pruning
6.0 Actual Resources Used
Sr. Name of Resources / Materials Specification Qty Remarks
No.

1 Computer RAM minimum 1

4GB,i5 7th Gen
2 Operating system Windows10 1

3 Internet Google 1

 What is Cloud Computing in Machine Learning.

Cloud computing in machine learning (ML) refers to the use of cloud-based

infrastructure, platforms, and services to build, deploy, and manage machine
learning models. Cloud computing provides a scalable, on-demand, and cost-
effective way to access computing resources, storage, and specialized hardware,
such as graphics processing units (GPUs) and tensor processing units (TPUs),
necessary for machine learning workloads.

Benefits of Cloud Computing in Machine Learning:

1. Scalability: Cloud computing resources can be scaled up or down to match

changing ML workload requirements.
2. Cost-effectiveness: Pay-as-you-go pricing models reduce costs compared to
maintaining on-premises infrastructure.
3. Faster deployment: Cloud services provide pre-configured environments and
tools for rapid ML development.
4. Access to specialized hardware: Cloud providers offer specialized hardware like
GPUs and TPUs, accelerating ML computations.
5. Collaboration: Cloud-based platforms enable seamless collaboration among data
scientists and developers.

 Diagram:-
o Advantages of cloud computing in machine learning:-

1. Scalability
2. Cost-effectiveness
3. Faster processing
4. Enhanced collaboration
5. Increased storage

o Disadvantages:-
1. Data dependency on internet
2. Security and privacy concerns
3. Vendor lock-in
4. Limited control over infrastructure

 Conclusion:-
Cloud computing makes machine learning faster, cheaper, and more
collaborative. However, it requires careful consideration of security, control, and
compliance. As technology advances, cloud-based machine learning will become
increasingly important for businesses and industries.

Cloud computing in machine learning combines the power of cloud computing

and machine learning to enable faster, more scalable, and cost-effective
processing of large datasets.
 Reference:-
www.google.com

www.wikepedia.com

www.greekgorgreeks.com

www.javatpoint.com

www.tutorialspoint.com
  Conclusion :-
The cloud computing models used in machine learning include public clouds
(such as AWS and Google Cloud), private clouds (on-premises), and hybrid
clouds (a combination of both). Popular machine learning cloud platforms
include AWS SageMaker, Google Cloud AI Platform, Microsoft Azure Machine
Learning, and IBM Cloud Pak for AI.
 MICRO-PROJECT EVOLUTION SHEET

Name of Student :- Nikita Shankar Hingmire En. No:-2210920387

Name of Program:- Semester:- 5th Sem

Course Name:-CLOUD COMPUTING FOR DATA SCIENCE.

Course Code:- 22594

Title of The Micro-Project:- CLOUD COMPUTING IN MACHINE LEARNING.

Course Outcomes Achieved:-

1. Deploy machine learning models on cloud platforms
2. Use cloud-based machine learning services (e.g., AWS SageMaker)
3. Manage large datasets in cloud storage
4. Scale machine learning workloads in cloud environments
5. Implement cloud-based data pipelines for machine learning
Sr. Characteristic to Poor Average Good Excellent Sub
(Marks1-3) (Marks4-5) (Marks 6-8) (Marks9-10) Total
No. be assessed
(A) Process and Product Assessment (Convert Below total marks out of 6Marks)
1 Relevance to the course
2 Literature
Review/informatio
n
collection
3 Completion of the
Target as Per project
proposal
4 Analysis of Data and
representation
5
Quality of
Prototype/Model
6 Report Preparation
(B) Individual Presentation/Viva(Convert Below total marks out of 4Marks)
7 Presentation
8
Viva

(A) (B)
Process and Product Individual Presentation/ Total Marks
Assessment (6 marks) Viva (4 marks) 10

Comments/Suggestions about team work/leadership/inter-personal communication (if any)

Name of Subject Teacher:-T.A.GAVHAL
Dated Signature:-