[go: up one dir, main page]
Skip to content
/ NPACS Public

Implementation of "Parameter Continuation Methods for the Optimization of Deep Neural Networks" accepted at ICMLA 19. https://ieeexplore.ieee.org/abstract/document/8999318

digitalcommons.wpi.edu/etd-theses/1256/
0 stars 0 forks Branches Tags Activity
Star
Notifications You must be signed in to change notification settings

harsh306/NPACS

BranchesTags

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
arc
arc
 
 
utils
utils
 
 
README.md
README.md
 
 
ae.py
ae.py
 
 
example.json
example.json
 
 
mlp.py
mlp.py
 
 
npc.py
npc.py
 
 
npc_adap_mlp.py
npc_adap_mlp.py
 
 
npc_sigmoid_grad.py
npc_sigmoid_grad.py
 
 
npcs.py
npcs.py
 
 
ops2.py
ops2.py
 
 

Repository files navigation

File ae.py contains the modal and TF graph.

File npc.py contains Natural parameter continuation (NPC) method (without secant).

File main.py contains NPCS method with secant.

File ops2.py contains basic operations.

File example.json has list of hyperparameter to tune.

Folder csmote has the code for Continuous-SMOTE.

About

Implementation of "Parameter Continuation Methods for the Optimization of Deep Neural Networks" accepted at ICMLA 19. https://ieeexplore.ieee.org/abstract/document/8999318

digitalcommons.wpi.edu/etd-theses/1256/

Topics

training deep-learning optimization continuation secant adam-optimizer autoen continat

Resources

Readme
Activity

Stars

0 stars

Watchers

2 watching

Forks

0 forks
Report repository

Releases

No releases published

Packages

No packages published

Languages