marvinzh / cma_es Public

Notifications You must be signed in to change notification settings
Fork 1
Star 0

a CMA-ES based hyperparameter optimization tool for NMT.

BSD-3-Clause license

0 stars 1 fork Branches Tags Activity

Notifications

Name		Name	Last commit message	Last commit date
Latest commit History 16 Commits
LICENSE		LICENSE
README.md		README.md
clean.sh		clean.sh
evo_single.py		evo_single.py
gene2conf.py		gene2conf.py
init.conf		init.conf
logger.py		logger.py
reporter.py		reporter.py
run.sh		run.sh
template.py		template.py
toy_nmt.py		toy_nmt.py

Repository files navigation

CMA-ES

a CMA-ES based hyperparameter optimization tool.

Prerequisite

pip install cma

basic usage:

./run.sh 0

Usage

evo_single.py : load initial hyperparameter or previous checkpoint, generate genes for current generation.

gene2conf.py: load genes for current generation, convert them to model description file.

reporter.py: load the evalution result file (e.g. BLEU score) output by each model. report them in a file.

toy_nmt.py: Script to generate random score as a fake evaluation score for testing. (format from sockeye-recipe)

About

a CMA-ES based hyperparameter optimization tool for NMT.

deeplearning hyperparameters-optimization

BSD-3-Clause license

Report repository

Releases

No releases published

Packages

No packages published

Languages