Slide 1
Slide 1 text
No content
Slide 2
Slide 2 text
Sunhyeong hong (홍선형)
What to eat
Today
MLU
Beginner
planner
Golf
Learning English
Travel
Platform Product
Manager
What to
drink
Today
Slide 3
Slide 3 text
Intro
Jason Allen's art created on Midjourney titled “Space Opera Theater”
Slide 4
Slide 4 text
Intro
Rain
Go Out Stay In
Too Hot Cable Signal?
Shopping Movie Coffee Shop TV Shows
Mode to predict Hong’s Weekend Plan
No Yes
Slide 5
Slide 5 text
Intro
Is it enough? ML-Model
development
Slide 6
Slide 6 text
Intro
Data sourcing
Streaming
Sourcing
Ingestion
Data
management
Validation
Analysis
Segmentation
Feature
engineering
ML-Model
development
Building
Training
Evaluation
Versioning
metadata
Application
Model
deployment
Model monitoring
Serving API/UI
Load balancing
Infrastructure
Configuration Containerization
Logging
monitoring
CI/CD
Pipelines
Authentication Hardware/GPUs
No. you have to do all this
Slide 7
Slide 7 text
Intro
These are the skills you need to do ML
Business
Programming Statices, ML/AI
Communication
The business
guy
The “if”
guy
The hacker
The number
cruncher
The data
translator
The ML/AI
engineer
The perfect
Data
scientist
The
salesperson
The Data
nerd
The
statistician The
Storyteller
The good
consultant
The competence
Science
professor
Slide 8
Slide 8 text
Intro
These are the skills you need to do ML
Business
Programming Statices, ML/AI
Communication
Are you okay...?
Slide 9
Slide 9 text
Agenda
- What is Machine Learning Pipeline?
- How to Build an End-to-End a Machine
Learning Pipeline?
- Building an End-to-End Pipeline using MLU
- Next MLU?
Slide 10
Slide 10 text
What is Machine Learning Pipeline?
Slide 11
Slide 11 text
How to Build an End-to-End a Machine
Learning Pipeline?
Data Extraction
and Analysis
Data
Preparation
Model Training
Model Evaluation
and Validation
Trained
Model
Manual experiment steps
Offline data
Manual ML Pipeline
Slide 12
Slide 12 text
How to Build an End-to-End a Machine
Learning Pipeline?
Data Extraction
and Analysis
Data
Preparation
Model Training
Model Evaluation
and Validation
Model Serving
Trained
Model
Manual experiment steps
Offline data
Model registry
Ops
Manual ML Pipeline
ML
Slide 13
Slide 13 text
How to Build an End-to-End a Machine
Learning Pipeline?
It's a new model
ML Ops
Slide 14
Slide 14 text
How to Build an End-to-End a Machine
Learning Pipeline?
It's a new model
I checked, but It doesn’t work
ML Ops
Slide 15
Slide 15 text
How to Build an End-to-End a Machine
Learning Pipeline?
It's a new model
What? I have no problem
I checked, but It doesn’t work
ML Ops
Slide 16
Slide 16 text
How to Build an End-to-End a Machine
Learning Pipeline?
It's a new model
What? I have no problem
I checked, but It doesn’t work
what python do you use?
ML Ops
Slide 17
Slide 17 text
How to Build an End-to-End a Machine
Learning Pipeline?
It's a new model
What? I have no problem
Ah, My Python version
is not compatible
, so I uploaded it to 3.8.
I checked, but It doesn’t work
what python do you use?
ML Ops
Slide 18
Slide 18 text
How to Build an End-to-End a Machine
Learning Pipeline?
It's a new model
What? I have no problem
Ah, My Python version
is not compatible
, so I uploaded it to 3.8.
I checked, but It doesn’t work
what python do you use?
Oh…
Please send requirements.txt
ML Ops
Slide 19
Slide 19 text
How to Build an End-to-End a Machine
Learning Pipeline?
Automation ML pipeline
Data
Extraction
Data
Preparation
Model
Training
Model
Evaluation
Model
Serving
Trained
Model
Code Repository
Model registry
Model
Monitoring
Data
Validation
Model
Validation
Feature
Store
Data Analysis
And Experimentation
Now
Code
Model registry
Slide 20
Slide 20 text
How to Build an End-to-End a Machine
Learning Pipeline?
Automation ML pipeline
Data
Extraction
Data
Preparation
Model
Training
Model
Evaluation
Model
Serving
Trained
Model
Automated pipeline
Code Repository
Model registry
Model
Monitoring
Data
Validation
Model
Validation
Feature
Store
Data Analysis
And Experimentation
Now
Code
Model registry
Slide 21
Slide 21 text
How to Build an End-to-End a Machine
Learning Pipeline?
Automation ML pipeline
Data
Extraction
Data
Preparation
Model
Training
Model
Evaluation
Model
Serving
Trained
Model
Automated pipeline
Code Repository
Model registry
Model
Monitoring
Data
Validation
Model
Validation
Feature
Store
Data Analysis
And Experimentation
Now
Code
Model registry
Slide 22
Slide 22 text
LINE's MLOps platform
Machine Learning Universe
Slide 23
Slide 23 text
Building an End-to-End Pipeline using MLU
VOOM For you recommendation system
Slide 24
Slide 24 text
Machine Learning Model
Building an End-to-End Pipeline using MLU
VOOM For you recommendation system
VOOM public Post
Create a
recommended
candidate group
Recommendation
candidate data
filtering and quality
verification
Recommendation
model Training
Slide 25
Slide 25 text
Machine Learning Model
Building an End-to-End Pipeline using MLU
VOOM For you recommendation system
Create a
recommended
candidate group
Recommendation
candidate data
filtering and quality
verification
Recommendation
model Training
Recommendation
model deployment
VOOM public Post
Recommendation
model monitoring
Slide 26
Slide 26 text
Machine Learning Model
Building an End-to-End Pipeline using MLU
VOOM For you recommendation system
Create a
recommended
candidate group
Recommendation
candidate data
filtering and quality
verification
Recommendation
model Training
Recommendation
model deployment
VOOM public Post
User Feedback
Recommendation
model monitoring
Slide 27
Slide 27 text
Continuous Data and Model updates
Slide 28
Slide 28 text
ML Model Logging and Tracking
Slide 29
Slide 29 text
ML Model API Packaging
Slide 30
Slide 30 text
Server resource optimization
Slide 31
Slide 31 text
Different development environment
Slide 32
Slide 32 text
MLU
Slide 33
Slide 33 text
Building an End-to-End Pipeline using MLU
The VOOM recommendation model process within the MLU architecture
Data
preprocessing
Model training
Feature
Store
Data
Interactive Computing Mountable Filesystem
Model
validation
Slide 34
Slide 34 text
Building an End-to-End Pipeline using MLU
The VOOM recommendation model process within the MLU architecture
Data
preprocessing
Model training
Feature
Store
Data
Interactive Computing
Experiment Management
Mountable Filesystem
Model
validation
Slide 35
Slide 35 text
Building an End-to-End Pipeline using MLU
The VOOM recommendation model process within the MLU architecture
Data
preprocessing
Model training
Feature
Store
Data
Interactive Computing
Experiment Management
Mountable Filesystem
Model
validation
Model
packaging
Model
deployment
Packaging Management
Slide 36
Slide 36 text
Building an End-to-End Pipeline using MLU
The VOOM recommendation model process within the MLU architecture
Data
preprocessing
Model training
Feature
Store
Data
Interactive Computing
Experiment Management
Mountable Filesystem
Model
validation
Model
packaging
Model
deployment
Workflow Management
Packaging Management
Slide 37
Slide 37 text
Building an End-to-End Pipeline using MLU
The VOOM recommendation model process within the MLU architecture
Data
preprocessing
Model training
Feature
Store
Data
Interactive Computing
Experiment Management
Mountable Filesystem
Model
validation
Model
packaging
Model
deployment
Workflow Management
Packaging Management
Model Serving
Model
Monitoring
Serving Management
Slide 38
Slide 38 text
Building an End-to-End Pipeline using MLU
The dashboard of the MLU portal
Slide 39
Slide 39 text
Building an End-to-End Pipeline using MLU
Monitoring screen of MLU serving
Slide 40
Slide 40 text
Building an End-to-End Pipeline using MLU
Pipeline operation accidents
90%↓
100%↓
! 95%↓
!
Slide 41
Slide 41 text
Building an End-to-End Pipeline using MLU
Improve model performance CTR 25% → 38% (13%↑)
Pipeline operation accidents
90%↓
100%↓
! 95%↓
!
Slide 42
Slide 42 text
Building an End-to-End Pipeline using MLU
Provisioning
Slide 43
Slide 43 text
Provisioning ML Ecosystem Minimal Code
Building an End-to-End Pipeline using MLU
Slide 44
Slide 44 text
Building an End-to-End Pipeline using MLU
Provisioning ML Ecosystem GUI Platform
Slide 45
Slide 45 text
Building an End-to-End Pipeline using MLU
MLU Usage Status - As of October ‘22
Cluster
GPU: 504 Cores
CPU: 14,886 Cores
MEM: 52TB
Slide 46
Slide 46 text
Building an End-to-End Pipeline using MLU
MLU Usage Status - As of October ‘22
Cluster
GPU: 504 Cores
CPU: 14,886 Cores
MEM: 52TB
User
Total Count: 1,111
Company: 33
Team: 386
Slide 47
Slide 47 text
Building an End-to-End Pipeline using MLU
MLU Usage Status - As of October ‘22
Cluster
GPU: 504 Cores
CPU: 14,886 Cores
MEM: 52TB
User
Total Count: 1,111
Company: 33
Team: 386
Pipeline
Pipeline : 300
Serving : 107
Slide 48
Slide 48 text
- Efficient Data management tools
- An active community where everyone can share their knowledge
- A Public repository for sharing Models and Datasets
Next MLU
Slide 49
Slide 49 text
- Efficient Data management tools
- An active community where everyone can share their knowledge
- A Public repository for sharing Models and Datasets
Next MLU
Slide 50
Slide 50 text
Released just yesterday
MLU MARKET PLACE
Slide 51
Slide 51 text
No content
Slide 52
Slide 52 text
ML Engineer
MLOps
Slide 53
Slide 53 text
ML Engineer
Marketer
Planner
Service Engineer
Designer
Server Engineer
MLU
Slide 54
Slide 54 text
Thank you