The dependencies include:
- python: 3.6.5
- gpy: 1.9.8
- gpflow: 1.4.1
- tensorflow: 1.14.0
- tensorflow-probability
python levelsetestimation.py --function $FUNCTION --criterion $CRITERION --level 0 --numqueries $NQUERIES --numruns 30 --numhyps 1 --noisevar $NOISEVAR --nparal 1 --ntrain 500 --nysample 5000 --ninit 2
where $FUNCTION are from functions.py; $CRITERION can be bes, straddle, dare; $NOISEVAR can be 0.0001, 0.09; $NQUERIES can be 100, 200.
python bayesianoptimization.py --function $FUNCTION --criterion $CRITERION --numqueries $NQUERIES --numruns 15 --numhyps 1 --noisevar $NOISEVAR --nmax 5 --nfeature 300 --nparal 2 --nsto 10 --ntrain 500 --nysample 3000 --ninit 2
where $FUNCTION are from functions.py; $CRITERION can be avg_bes_mp (BES-MP), pes, ucb, ei, mes; $NOISEVAR can be 0.0001, 0.09; $NQUERIES can be 100, 200.
python implicitlse.py --function $FUNCTION --criterion $CRITERION --alpha 0.2 --numqueries 200 --numruns 30 --numhyps 1 --nmax 5 --nfeature 300 --noisevar 0.0001 --nparal 1 --ntrain 500 --nysample 3000 --ninit 2
where $FUNCTION are from functions.py; $CRITERION can be mnes2 (BES-MP), mnes3 (BES^2-MP).
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent