Agendendations of

D ML
engriering

E
2) Deep Leaving
tools
(Roadmap) GenAl -

3)
Lifecycle
ML
a

1)
5)
ML

MLOPS
-
System Design Wales co , und
d

learn
Resources to
-

S Machining Engriering
Travi a model

on historical

t production
Scalable ,
reliable , optimised

productmanage
a

Date Sulutists , softwaregnie ,

skills probability & stats Calculus

u Algebra
,
Linear
-
,

Programing EPython'
-
 leaving
essentials
2) Machine
model on historical
*
Supervised learning
+ Train
date
=

E
+

labels

I
-

but historical date

labels
Unsupervised tearing
- No

fR -
means Clustering

to
Reduction
Divisionality

3)Tools
librarya upyter Morbok ,age
/ was

Libraries If Datedaiq -
Pandas
Data Vizs Seabor
Sci-kit lear , Xost
leaving
-

Machine

Deep Larry -

-
Tenoflow/ Pyra
↓

google facebook

-Large
Language (library)
r

" Pytor # Facebook

 /Deops

I I
An a

Gitdita
,

Code
=
Outia Serve
I -

colaboration Git Github

hosting E
or
-
code

i
-

-
a
(button

Project
-
lode is
ready (hinc) I

= state
Dockerize/Contanicize
notE Container

#garden
-

Service

a
is
Services of Orchestrator

tan As -

Host this https://

aws.amazon.com/ecs/
O ECR
-

https://www.docker.com/ Elastic Container

Regi,sd)
te
https://aws.amazon.com/ecr/

&
Deep Machin
armig Lang
↓ Image recognition peep
hearing
tINLP
↓ Recommendation Systei
↓ This
Series Analysis

Popular
Architectures

Neural Networks
CNN - Convolution -

detection
10 object
&

10) 256

X 256

RWNs ,
LSTM ·

3 Time series

architecture
Transforme (2017)

E a
and
"Attention
Iboughta des
my
color
t is white is
 Chat

Game
Transforma
Generative

open

n
ML
Lifecycle
project
1) Gathes date

2) Explore -

Exploratory
date
are
lysis
Pandas ,
seaborn

3)
featureenquiring
supervised stikitteam

4)
buildingsuprise
Me
-Langhai
Tensorflow / Bylorch

37 Aralmation ME (skram classificat

Accuracy
-

,
Flscore ,
Metrics)
-
6) (Deploy +
Docke - Ars (E2s)
 Project
code https://github.com/aws-actions/amazon-ecs-deploy-task-
definition

# Gitation
Histt
aws
Gitm E
A

dockerize
a
project
l
n

change
is some
If there

I
a

onmai
it
merge

·
mmm Aus
↓
connection

stu

Es

Ip
public-
Jame

address
(and)

(auto)
 LogginW
g hat
experiments fwhile buildingAm model)

↓
SystemDesign

Cass 24
* https://aws.amazon.com/s3/

Database
website

·
↑

https://aws.amazon.com/
redshift/
Basic implementation
xaboose f
other
algos
of
https://colab.research.google.com/drive/13rBvKSSjne8lfnpvxQmWFf6-bs6mI_ft?usp=sharing

with Code
End-to end 21/CD pipeline
https://github.com/gitsuraaj/DSML_KITTY_FLASK/tree/main