How many iterations for model-based optimization (in mlrMBO) are necessary?
up vote
5
down vote
favorite
I would like to use model-based optimization within the mlr-Package in R (mlrMBO) to tune my hyperparameters.
How many iterations are recommended here?
I have read that the number of necessary iterations in MBO depends on the number of hyperparameters, that should be multiplied by a certain factor?
r performance hyperparameters mlr
add a comment |
up vote
5
down vote
favorite
I would like to use model-based optimization within the mlr-Package in R (mlrMBO) to tune my hyperparameters.
How many iterations are recommended here?
I have read that the number of necessary iterations in MBO depends on the number of hyperparameters, that should be multiplied by a certain factor?
r performance hyperparameters mlr
add a comment |
up vote
5
down vote
favorite
up vote
5
down vote
favorite
I would like to use model-based optimization within the mlr-Package in R (mlrMBO) to tune my hyperparameters.
How many iterations are recommended here?
I have read that the number of necessary iterations in MBO depends on the number of hyperparameters, that should be multiplied by a certain factor?
r performance hyperparameters mlr
I would like to use model-based optimization within the mlr-Package in R (mlrMBO) to tune my hyperparameters.
How many iterations are recommended here?
I have read that the number of necessary iterations in MBO depends on the number of hyperparameters, that should be multiplied by a certain factor?
r performance hyperparameters mlr
r performance hyperparameters mlr
edited Nov 10 at 4:12
Mozahler
2,03641625
2,03641625
asked Oct 31 at 13:01
Ana
263
263
add a comment |
add a comment |
1 Answer
1
active
oldest
votes
up vote
2
down vote
Unfortunately there's no general answer to this. In addition to the number of hyperparameters, the difficulty of the task and the general optimization landscape have an impact on how many iterations may be needed. Furthermore, it depends on what level of performance you're looking for -- if you absolutely need the best performance, you may need a lot of iterations (thousands or more).
A good rule of thumb is to determine the number of iterations based on the resources you have available. Do you need an answer in a few minutes or can you leave it run overnight? I would start with a small test run with only a few iterations to get a feeling for how long it takes in your particular case and then set the number of iterations for a real run based on that and how long you want to spend on it.
It's also a good idea to do multiple runs with different random seeds and the same number of iterations and the same random seed and different numbers of iterations. This will tell you how much more performance you can get by running longer, e.g. if there is no difference for 100 and 1000 iterations then giving it more may not be worth it.
add a comment |
Your Answer
StackExchange.ifUsing("editor", function () {
StackExchange.using("externalEditor", function () {
StackExchange.using("snippets", function () {
StackExchange.snippets.init();
});
});
}, "code-snippets");
StackExchange.ready(function() {
var channelOptions = {
tags: "".split(" "),
id: "1"
};
initTagRenderer("".split(" "), "".split(" "), channelOptions);
StackExchange.using("externalEditor", function() {
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled) {
StackExchange.using("snippets", function() {
createEditor();
});
}
else {
createEditor();
}
});
function createEditor() {
StackExchange.prepareEditor({
heartbeatType: 'answer',
convertImagesToLinks: true,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: 10,
bindNavPrevention: true,
postfix: "",
imageUploader: {
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
},
onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
});
}
});
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f53083989%2fhow-many-iterations-for-model-based-optimization-in-mlrmbo-are-necessary%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
1 Answer
1
active
oldest
votes
1 Answer
1
active
oldest
votes
active
oldest
votes
active
oldest
votes
up vote
2
down vote
Unfortunately there's no general answer to this. In addition to the number of hyperparameters, the difficulty of the task and the general optimization landscape have an impact on how many iterations may be needed. Furthermore, it depends on what level of performance you're looking for -- if you absolutely need the best performance, you may need a lot of iterations (thousands or more).
A good rule of thumb is to determine the number of iterations based on the resources you have available. Do you need an answer in a few minutes or can you leave it run overnight? I would start with a small test run with only a few iterations to get a feeling for how long it takes in your particular case and then set the number of iterations for a real run based on that and how long you want to spend on it.
It's also a good idea to do multiple runs with different random seeds and the same number of iterations and the same random seed and different numbers of iterations. This will tell you how much more performance you can get by running longer, e.g. if there is no difference for 100 and 1000 iterations then giving it more may not be worth it.
add a comment |
up vote
2
down vote
Unfortunately there's no general answer to this. In addition to the number of hyperparameters, the difficulty of the task and the general optimization landscape have an impact on how many iterations may be needed. Furthermore, it depends on what level of performance you're looking for -- if you absolutely need the best performance, you may need a lot of iterations (thousands or more).
A good rule of thumb is to determine the number of iterations based on the resources you have available. Do you need an answer in a few minutes or can you leave it run overnight? I would start with a small test run with only a few iterations to get a feeling for how long it takes in your particular case and then set the number of iterations for a real run based on that and how long you want to spend on it.
It's also a good idea to do multiple runs with different random seeds and the same number of iterations and the same random seed and different numbers of iterations. This will tell you how much more performance you can get by running longer, e.g. if there is no difference for 100 and 1000 iterations then giving it more may not be worth it.
add a comment |
up vote
2
down vote
up vote
2
down vote
Unfortunately there's no general answer to this. In addition to the number of hyperparameters, the difficulty of the task and the general optimization landscape have an impact on how many iterations may be needed. Furthermore, it depends on what level of performance you're looking for -- if you absolutely need the best performance, you may need a lot of iterations (thousands or more).
A good rule of thumb is to determine the number of iterations based on the resources you have available. Do you need an answer in a few minutes or can you leave it run overnight? I would start with a small test run with only a few iterations to get a feeling for how long it takes in your particular case and then set the number of iterations for a real run based on that and how long you want to spend on it.
It's also a good idea to do multiple runs with different random seeds and the same number of iterations and the same random seed and different numbers of iterations. This will tell you how much more performance you can get by running longer, e.g. if there is no difference for 100 and 1000 iterations then giving it more may not be worth it.
Unfortunately there's no general answer to this. In addition to the number of hyperparameters, the difficulty of the task and the general optimization landscape have an impact on how many iterations may be needed. Furthermore, it depends on what level of performance you're looking for -- if you absolutely need the best performance, you may need a lot of iterations (thousands or more).
A good rule of thumb is to determine the number of iterations based on the resources you have available. Do you need an answer in a few minutes or can you leave it run overnight? I would start with a small test run with only a few iterations to get a feeling for how long it takes in your particular case and then set the number of iterations for a real run based on that and how long you want to spend on it.
It's also a good idea to do multiple runs with different random seeds and the same number of iterations and the same random seed and different numbers of iterations. This will tell you how much more performance you can get by running longer, e.g. if there is no difference for 100 and 1000 iterations then giving it more may not be worth it.
answered Oct 31 at 15:08
Lars Kotthoff
90k11157164
90k11157164
add a comment |
add a comment |
Thanks for contributing an answer to Stack Overflow!
- Please be sure to answer the question. Provide details and share your research!
But avoid …
- Asking for help, clarification, or responding to other answers.
- Making statements based on opinion; back them up with references or personal experience.
To learn more, see our tips on writing great answers.
Some of your past answers have not been well-received, and you're in danger of being blocked from answering.
Please pay close attention to the following guidance:
- Please be sure to answer the question. Provide details and share your research!
But avoid …
- Asking for help, clarification, or responding to other answers.
- Making statements based on opinion; back them up with references or personal experience.
To learn more, see our tips on writing great answers.
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f53083989%2fhow-many-iterations-for-model-based-optimization-in-mlrmbo-are-necessary%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown