Accenture Enterprise AI – Scaling Machine Learning and Deep Learning Models
Table of Contents
Abstract and
introduction
...................................................................................................................................i
Abstract
.............................................................................................................................................................
1
Are you Well-Architected?
............................................................................................................................
1
Introduction
......................................................................................................................................................
2
Frictionless ideation to production
..............................................................................................................
2
Workforce analytics use cases........................................................................................................................3
An intelligent platform approach
.................................................................................................................
4
ML architecture on AWS...................................................................................................................................5
Feature engineering ...........................................................................................................................................6
The feature store
............................................................................................................................................
6
The algorithms.....................................................................................................................................................8
Data engineering and data quality
..............................................................................................................
8
Hyper-parameter tuning (HPT)
...................................................................................................................
8
Model registry
...............................................................................................................................................
10
Optimization
......................................................................................................................................................
11
Optimization drivers
....................................................................................................................................
11
Fine-tuning and reuse of models
..............................................................................................................
11
Scaling with distributed training
...............................................................................................................
13
Avoiding common missteps to reduce
rework........................................................................................
13
Machine learning pipelines ............................................................................................................................15
Going from POC to large-scale deployments
.........................................................................................
15
Applying software engineering principles to data science
............................................................
16
Machine learning automation through
pipelines ...................................................................................
17
Tracking
lineage
............................................................................................................................................
18
Monitoring for performance and bias ........................................................................................................19
Post-training bias metrics
...........................................................................................................................
20
Monitoring performance
.............................................................................................................................
20
Data quality monitoring
.........................................................................................................................
20
VIP享7折，此内容立减5.97元 
