How to optimize NN seq2seq classification training with time-dependent class imbalance?
$begingroup$
I want to classify timeseries using seq2seq since they have a high chance of being "corrupted" towards the end.
The traces are labelled e.g.
111111100000
555500000000
330000000000
444444440000
444444444444
as you can see, they (almost) always end up in zero-land.
If I plot the label-distribution over time, it looks something like the plot below, where the trace has an increasingly growing chance of being terminated and zero-ed at some point.
Are there any common methods for balancing problems like this, or should I simply try to balance classes based on e.g. the first n datapoints only?
neural-network deep-learning classification sequence-to-sequence
asked yesterday
komodovaran_komodovaran_
1114
$endgroup$
add a comment |
$begingroup$
I want to classify timeseries using seq2seq since they have a high chance of being "corrupted" towards the end.
The traces are labelled e.g.
111111100000
555500000000
330000000000
444444440000
444444444444
as you can see, they (almost) always end up in zero-land.
If I plot the label-distribution over time, it looks something like the plot below, where the trace has an increasingly growing chance of being terminated and zero-ed at some point.
Are there any common methods for balancing problems like this, or should I simply try to balance classes based on e.g. the first n datapoints only?
neural-network deep-learning classification sequence-to-sequence
asked yesterday
komodovaran_komodovaran_
1114
$endgroup$
add a comment |
$begingroup$
I want to classify timeseries using seq2seq since they have a high chance of being "corrupted" towards the end.
The traces are labelled e.g.
111111100000
555500000000
330000000000
444444440000
444444444444
as you can see, they (almost) always end up in zero-land.
If I plot the label-distribution over time, it looks something like the plot below, where the trace has an increasingly growing chance of being terminated and zero-ed at some point.
Are there any common methods for balancing problems like this, or should I simply try to balance classes based on e.g. the first n datapoints only?
neural-network deep-learning classification sequence-to-sequence
asked yesterday
komodovaran_komodovaran_
1114
$endgroup$
I want to classify timeseries using seq2seq since they have a high chance of being "corrupted" towards the end.
The traces are labelled e.g.
111111100000
555500000000
330000000000
444444440000
444444444444
as you can see, they (almost) always end up in zero-land.
If I plot the label-distribution over time, it looks something like the plot below, where the trace has an increasingly growing chance of being terminated and zero-ed at some point.
Are there any common methods for balancing problems like this, or should I simply try to balance classes based on e.g. the first n datapoints only?
neural-network deep-learning classification sequence-to-sequence
neural-network deep-learning classification sequence-to-sequence
asked yesterday
komodovaran_komodovaran_
1114
asked yesterday
komodovaran_komodovaran_
1114
asked yesterday
komodovaran_komodovaran_
1114
asked yesterday
komodovaran_komodovaran_
1114
asked yesterday
komodovaran_komodovaran_
1114
1114
add a comment |
add a comment |
0
active
oldest
votes
Your Answer
StackExchange.ifUsing("editor", function () {
return StackExchange.using("mathjaxEditing", function () {
StackExchange.MarkdownEditor.creationCallbacks.add(function (editor, postfix) {
StackExchange.mathjaxEditing.prepareWmdForMathJax(editor, postfix, [["$", "$"], ["\\(","\\)"]]);
});
});
}, "mathjax-editing");
StackExchange.ready(function() {
var channelOptions = {
tags: "".split(" "),
id: "557"
};
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',
autoActivateHeartbeat: false,
convertImagesToLinks: false,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: null,
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%2fdatascience.stackexchange.com%2fquestions%2f47548%2fhow-to-optimize-nn-seq2seq-classification-training-with-time-dependent-class-imb%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
0
active
oldest
votes
0
active
oldest
votes
active
oldest
votes
active
oldest
votes
Thanks for contributing an answer to Data Science Stack Exchange!
- 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.
Use MathJax to format equations. MathJax reference.
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%2fdatascience.stackexchange.com%2fquestions%2f47548%2fhow-to-optimize-nn-seq2seq-classification-training-with-time-dependent-class-imb%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
