Difference in accuracy in FC-DNN on different run











up vote
0
down vote

favorite












I have been facing an issue of instability in deep neural networks. For example, while using DNN with fully connected layers (3 layers, 128 memory units in each), I am having a range of unweighted accuracy of 62% to 66% in the different runs (yes, the same train and test set). I used Xavier initializer and bias initializer as zero. All the other parameters are also fixed.



Have anyone faced such problems? I have tried to use fixed seed too, but it doesn't seem to help. Kinda puzzled about what is creating this instability.



Thanks in advance :)






neural-network deep-learning







share|improve this question






















  • This is completely normal, and its due to the random weight initialization.
    – Matias Valdenegro
    yesterday















up vote
0
down vote

favorite












I have been facing an issue of instability in deep neural networks. For example, while using DNN with fully connected layers (3 layers, 128 memory units in each), I am having a range of unweighted accuracy of 62% to 66% in the different runs (yes, the same train and test set). I used Xavier initializer and bias initializer as zero. All the other parameters are also fixed.



Have anyone faced such problems? I have tried to use fixed seed too, but it doesn't seem to help. Kinda puzzled about what is creating this instability.



Thanks in advance :)






neural-network deep-learning







share|improve this question






















  • This is completely normal, and its due to the random weight initialization.
    – Matias Valdenegro
    yesterday













up vote
0
down vote

favorite









up vote
0
down vote

favorite











I have been facing an issue of instability in deep neural networks. For example, while using DNN with fully connected layers (3 layers, 128 memory units in each), I am having a range of unweighted accuracy of 62% to 66% in the different runs (yes, the same train and test set). I used Xavier initializer and bias initializer as zero. All the other parameters are also fixed.



Have anyone faced such problems? I have tried to use fixed seed too, but it doesn't seem to help. Kinda puzzled about what is creating this instability.



Thanks in advance :)






neural-network deep-learning







share|improve this question













I have been facing an issue of instability in deep neural networks. For example, while using DNN with fully connected layers (3 layers, 128 memory units in each), I am having a range of unweighted accuracy of 62% to 66% in the different runs (yes, the same train and test set). I used Xavier initializer and bias initializer as zero. All the other parameters are also fixed.



Have anyone faced such problems? I have tried to use fixed seed too, but it doesn't seem to help. Kinda puzzled about what is creating this instability.



Thanks in advance :)






neural-network deep-learning




neural-network deep-learning






share|improve this question













share|improve this question











share|improve this question




share|improve this question










asked yesterday









Albert

1




1












  • This is completely normal, and its due to the random weight initialization.
    – Matias Valdenegro
    yesterday


















  • This is completely normal, and its due to the random weight initialization.
    – Matias Valdenegro
    yesterday
















This is completely normal, and its due to the random weight initialization.
– Matias Valdenegro
yesterday




This is completely normal, and its due to the random weight initialization.
– Matias Valdenegro
yesterday

















active

oldest

votes











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
});


}
});














 

draft saved


draft discarded


















StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f53237890%2fdifference-in-accuracy-in-fc-dnn-on-different-run%23new-answer', 'question_page');
}
);

Post as a guest























active

oldest

votes













active

oldest

votes









active

oldest

votes






active

oldest

votes
















 

draft saved


draft discarded



















































 


draft saved


draft discarded














StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f53237890%2fdifference-in-accuracy-in-fc-dnn-on-different-run%23new-answer', 'question_page');
}
);

Post as a guest






































































-

Popular posts from this blog

Full-time equivalent

Image
Full-time equivalent From Wikipedia, the free encyclopedia Jump to navigation Jump to search Full-time equivalent ( FTE ) or whole time equivalent ( WTE ) is a unit that indicates the workload of an employed person (or student) in a way that makes workloads or class loads comparable [1] across various contexts. FTE is often used to measure a worker's or student's involvement in a project, or to track cost reductions in an organization. An FTE of 1.0 is equivalent to a full-time worker or student, while an FTE of 0.5 signals half of a full work or school load. [2] Contents 1 U.S. Federal Government 2 In education 2.1 Example 3 Notes 4 References U.S. Federal Government [ edit ] In the U.S. Federal Government, FTE is defined by the Government Accountability Office (GAO) as the number of total hours worked divided by the maximum number of compensable hours in a full-time schedule as
Read more

Bicuculline

Image
Bicuculline From Wikipedia, the free encyclopedia Jump to navigation Jump to search Bicuculline Clinical data ATC code none Identifiers IUPAC name (6 R )-​6-​[(5 S )-​6-​methyl-​5,​6,​7,​8-​tetrahydro​[1,3]dioxolo​[4,5- g ]​isoquinolin-​5-​yl]​furo[3,4- e ]​[1,3]​benzodioxol-​8(6 H )-​one CAS Number 485-49-4   Y PubChem CID 10237 IUPHAR/BPS 2312 ChemSpider 9820   Y UNII Y37615DVKC ChEBI CHEBI:3092   N ChEMBL ChEMBL417990   N ECHA InfoCard 100.006.927 Chemical and physical data Formula C 20 H 17 N O 6 Molar mass 367.352 g/mol 3D model (JSmol) Interactive image Melting point 215 °C (419 °F) SMILES O=C1O[C@H](c3c1c2OCOc2cc3)[C@@H]5c4cc6OCOc6cc4CCN5C InChI InChI=1S/C20H17NO6/c1-21-5-4-10-6-14-15(25-8-24-14)7-12(10)17(21)18-11-2-3-13-19(26-9-23-13)16(11)20(22)27-18/h2-3,6-7,17-18H,4-5,8-9H2,1H3/t17-,18+/m0/s1   Y Key:IYGYMKDQCDOMRE-ZWKOTPCHSA-N
Read more

さくらももこ

Image
さくら ももこ 本名 非公開 [1] 生誕 ( 1965-05-08 ) 1965年5月8日 日本・静岡県清水市 （現・静岡市清水区） 死没 ( 2018-08-15 ) 2018年8月15日（53歳没） 日本・東京都目黒区 国籍 日本 職業 漫画家 エッセイスト 作詞家 活動期間 1984年 - 2018年 ジャンル 少女漫画 青年漫画 児童漫画 代表作 『ちびまる子ちゃん』 『コジコジ』 受賞 1989年：第13回講談社漫画賞少女部門 （『ちびまる子ちゃん』） 1990年：第32回日本レコード大賞 （『おどるポンポコリン』作詞） 公式サイト さくらプロダクション テンプレートを表示 さくら ももこ （1965年5月8日 - 2018年8月15日 [2] ）は、日本の漫画家、エッセイスト、作詞家、脚本家。また、自身の少女時代をモデルとした代表作のコミック『ちびまる子ちゃん』の主人公の名前でもある。静岡県清水市（現・静岡市清水区）出身。血液型はA型。身長159cm。一男の母親。 代表作のコミック『ちびまる子ちゃん』の単行本の売上は累計3000万部を超える。また、エッセイストとしても独特の視点と語り口で人気が高く、初期エッセイ集三部作『もものかんづめ』『さるのこしかけ』『たいのおかしら』はいずれもミリオンセラーを記録。 目次 1 経歴 2 人物・交遊関係 3 主な作品 3.1 漫画 3.2 エッセイ 3.3 作詞 3.4 詩集 3.5 雑誌・ムック本 3.6 翻訳 3.7 絵本、ドラマ脚本他 3.8 ラジオ出演 3.9 その他 4 関係人物 4.1 アシスタント 4.2 さくらももこを演じた人物 5 脚注 5.1 注釈 5.2 出典 6 外部リンク 経歴 年譜形式の経歴は推奨されていません 。人物の伝記は流れのあるまとまった文章で記述し、年譜は補助的な使用にとどめてください。 （ 2018年8月 ） 清水市立（当時）入江小学校
Read more