Mixing
CNN training accuracy drop suddenly from 94% into 48%
I am working in audio classification using CNN. The raining accuracy goes normally in an increasing manner till it reaches (94.9%) in epoch (61), then all the sudden in epoch (62) the training accuracy dropped to (48.37%). I don't know what is the problem that causes this.
37792/37792 [===========] - 1056s 28ms/step - loss: 0.1544 - acc: 0.9490
Epoch 61/100
37792/37792 [==========] - 1056s 28ms/step - loss: 8.3078 - acc: 0.4837
Epoch 62/100
.......................................................
tensorflow keras deep-learning conv-neural-network
edited Jan 2 at 6:27
NIMISHAN
95011223
asked Jan 2 at 2:50
m.aldarwbim.aldarwbi
162
add a comment |
I am working in audio classification using CNN. The raining accuracy goes normally in an increasing manner till it reaches (94.9%) in epoch (61), then all the sudden in epoch (62) the training accuracy dropped to (48.37%). I don't know what is the problem that causes this.
37792/37792 [===========] - 1056s 28ms/step - loss: 0.1544 - acc: 0.9490
Epoch 61/100
37792/37792 [==========] - 1056s 28ms/step - loss: 8.3078 - acc: 0.4837
Epoch 62/100
.......................................................
tensorflow keras deep-learning conv-neural-network
edited Jan 2 at 6:27
NIMISHAN
95011223
asked Jan 2 at 2:50
m.aldarwbim.aldarwbi
162
2
This could be a data related issue, check if the training dataset has proper labels, besides how does it behave after 62nd epoch?
– Oswald
Jan 2 at 3:14
It behaves bad after 62 epoch.
– m.aldarwbi
Jan 5 at 2:33
add a comment |
I am working in audio classification using CNN. The raining accuracy goes normally in an increasing manner till it reaches (94.9%) in epoch (61), then all the sudden in epoch (62) the training accuracy dropped to (48.37%). I don't know what is the problem that causes this.
37792/37792 [===========] - 1056s 28ms/step - loss: 0.1544 - acc: 0.9490
Epoch 61/100
37792/37792 [==========] - 1056s 28ms/step - loss: 8.3078 - acc: 0.4837
Epoch 62/100
.......................................................
tensorflow keras deep-learning conv-neural-network
edited Jan 2 at 6:27
NIMISHAN
95011223
asked Jan 2 at 2:50
m.aldarwbim.aldarwbi
162
I am working in audio classification using CNN. The raining accuracy goes normally in an increasing manner till it reaches (94.9%) in epoch (61), then all the sudden in epoch (62) the training accuracy dropped to (48.37%). I don't know what is the problem that causes this.
37792/37792 [===========] - 1056s 28ms/step - loss: 0.1544 - acc: 0.9490
Epoch 61/100
37792/37792 [==========] - 1056s 28ms/step - loss: 8.3078 - acc: 0.4837
Epoch 62/100
.......................................................
tensorflow keras deep-learning conv-neural-network
tensorflow keras deep-learning conv-neural-network
edited Jan 2 at 6:27
NIMISHAN
95011223
asked Jan 2 at 2:50
m.aldarwbim.aldarwbi
162
edited Jan 2 at 6:27
NIMISHAN
95011223
asked Jan 2 at 2:50
m.aldarwbim.aldarwbi
162
edited Jan 2 at 6:27
NIMISHAN
95011223
edited Jan 2 at 6:27
NIMISHAN
95011223
edited Jan 2 at 6:27
NIMISHAN
95011223
95011223
asked Jan 2 at 2:50
m.aldarwbim.aldarwbi
162
asked Jan 2 at 2:50
m.aldarwbim.aldarwbi
162
asked Jan 2 at 2:50
m.aldarwbim.aldarwbi
162
162
2
This could be a data related issue, check if the training dataset has proper labels, besides how does it behave after 62nd epoch?
– Oswald
Jan 2 at 3:14
It behaves bad after 62 epoch.
– m.aldarwbi
Jan 5 at 2:33
add a comment |
2
This could be a data related issue, check if the training dataset has proper labels, besides how does it behave after 62nd epoch?
– Oswald
Jan 2 at 3:14
It behaves bad after 62 epoch.
– m.aldarwbi
Jan 5 at 2:33
2
2
This could be a data related issue, check if the training dataset has proper labels, besides how does it behave after 62nd epoch?
– Oswald
Jan 2 at 3:14
This could be a data related issue, check if the training dataset has proper labels, besides how does it behave after 62nd epoch?
– Oswald
Jan 2 at 3:14
It behaves bad after 62 epoch.
– m.aldarwbi
Jan 5 at 2:33
It behaves bad after 62 epoch.
– m.aldarwbi
Jan 5 at 2:33
add a comment |
1 Answer
1
active
oldest
votes
I think that I've just solved the problem by customising the default optimiser.lr = 0.0001
beta_1 = 0.9
beta_2 = 0.999
opti = keras.optimizers.adam(lr=lr, beta_1=beta_1, beta_2=beta_2, epsilon=None, clipnorm=1.)
answered Jan 10 at 1:27
m.aldarwbim.aldarwbi
162
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',
autoActivateHeartbeat: false,
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%2f54000697%2fcnn-training-accuracy-drop-suddenly-from-94-into-48%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
I think that I've just solved the problem by customising the default optimiser.lr = 0.0001
beta_1 = 0.9
beta_2 = 0.999
opti = keras.optimizers.adam(lr=lr, beta_1=beta_1, beta_2=beta_2, epsilon=None, clipnorm=1.)
answered Jan 10 at 1:27
m.aldarwbim.aldarwbi
162
add a comment |
I think that I've just solved the problem by customising the default optimiser.lr = 0.0001
beta_1 = 0.9
beta_2 = 0.999
opti = keras.optimizers.adam(lr=lr, beta_1=beta_1, beta_2=beta_2, epsilon=None, clipnorm=1.)
answered Jan 10 at 1:27
m.aldarwbim.aldarwbi
162
add a comment |
I think that I've just solved the problem by customising the default optimiser.lr = 0.0001
beta_1 = 0.9
beta_2 = 0.999
opti = keras.optimizers.adam(lr=lr, beta_1=beta_1, beta_2=beta_2, epsilon=None, clipnorm=1.)
answered Jan 10 at 1:27
m.aldarwbim.aldarwbi
162
I think that I've just solved the problem by customising the default optimiser.lr = 0.0001
beta_1 = 0.9
beta_2 = 0.999
opti = keras.optimizers.adam(lr=lr, beta_1=beta_1, beta_2=beta_2, epsilon=None, clipnorm=1.)
answered Jan 10 at 1:27
m.aldarwbim.aldarwbi
162
answered Jan 10 at 1:27
m.aldarwbim.aldarwbi
162
answered Jan 10 at 1:27
m.aldarwbim.aldarwbi
162
answered Jan 10 at 1:27
m.aldarwbim.aldarwbi
162
162
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.
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%2f54000697%2fcnn-training-accuracy-drop-suddenly-from-94-into-48%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
2
This could be a data related issue, check if the training dataset has proper labels, besides how does it behave after 62nd epoch?
– Oswald
Jan 2 at 3:14
It behaves bad after 62 epoch.
– m.aldarwbi
Jan 5 at 2:33