Keras Loss Value Extremely High + Prediction Result same
$begingroup$
All the Keras Deep Learning tutorials feature an already wrapped image dataset, and there's just one load method to load it all.
I have a set of images and a corresponding csv file for targets of those images. I can't use the Image Data Generator as it's a regression problem and not a labeling one. So I made a custom numpy array of the following:
a numpy array of (448, 448, 3) images
a numpy array of corresponding target numbers
When this is fed into the model, I face no error/ exception. Except the output looks ridiculously bad. The loss encountered is extremely high (in thousands), and the data really does not look so incoherent. Maybe it's the model. Here's the description:
Sequential with 2 Conv2D Layers (64, 32), flattened it to feed to a Dense layer of 16 and 8, and then one final Dense layer with 1 output node with no activation function (because, regression). [Also tried to scale the image values to [0,1], but no luck.]
As a noob in this domain, I have no idea where to begin to know where I could be going wrong. If it's the way I'm loading up the data, can anyone guide me to how to go about with that. Thanks.
Just Observed:
The target test values are in the range 1 - 15 with whole numbers as the target value. The predictions are all the same, with a value around 0.00065. What could be the reason for this behaviour?
neural-network deep-learning keras regression
New contributor
$endgroup$
add a comment |
$begingroup$
All the Keras Deep Learning tutorials feature an already wrapped image dataset, and there's just one load method to load it all.
I have a set of images and a corresponding csv file for targets of those images. I can't use the Image Data Generator as it's a regression problem and not a labeling one. So I made a custom numpy array of the following:
a numpy array of (448, 448, 3) images
a numpy array of corresponding target numbers
When this is fed into the model, I face no error/ exception. Except the output looks ridiculously bad. The loss encountered is extremely high (in thousands), and the data really does not look so incoherent. Maybe it's the model. Here's the description:
Sequential with 2 Conv2D Layers (64, 32), flattened it to feed to a Dense layer of 16 and 8, and then one final Dense layer with 1 output node with no activation function (because, regression). [Also tried to scale the image values to [0,1], but no luck.]
As a noob in this domain, I have no idea where to begin to know where I could be going wrong. If it's the way I'm loading up the data, can anyone guide me to how to go about with that. Thanks.
Just Observed:
The target test values are in the range 1 - 15 with whole numbers as the target value. The predictions are all the same, with a value around 0.00065. What could be the reason for this behaviour?
neural-network deep-learning keras regression
New contributor
$endgroup$
add a comment |
$begingroup$
All the Keras Deep Learning tutorials feature an already wrapped image dataset, and there's just one load method to load it all.
I have a set of images and a corresponding csv file for targets of those images. I can't use the Image Data Generator as it's a regression problem and not a labeling one. So I made a custom numpy array of the following:
a numpy array of (448, 448, 3) images
a numpy array of corresponding target numbers
When this is fed into the model, I face no error/ exception. Except the output looks ridiculously bad. The loss encountered is extremely high (in thousands), and the data really does not look so incoherent. Maybe it's the model. Here's the description:
Sequential with 2 Conv2D Layers (64, 32), flattened it to feed to a Dense layer of 16 and 8, and then one final Dense layer with 1 output node with no activation function (because, regression). [Also tried to scale the image values to [0,1], but no luck.]
As a noob in this domain, I have no idea where to begin to know where I could be going wrong. If it's the way I'm loading up the data, can anyone guide me to how to go about with that. Thanks.
Just Observed:
The target test values are in the range 1 - 15 with whole numbers as the target value. The predictions are all the same, with a value around 0.00065. What could be the reason for this behaviour?
neural-network deep-learning keras regression
New contributor
$endgroup$
All the Keras Deep Learning tutorials feature an already wrapped image dataset, and there's just one load method to load it all.
I have a set of images and a corresponding csv file for targets of those images. I can't use the Image Data Generator as it's a regression problem and not a labeling one. So I made a custom numpy array of the following:
a numpy array of (448, 448, 3) images
a numpy array of corresponding target numbers
When this is fed into the model, I face no error/ exception. Except the output looks ridiculously bad. The loss encountered is extremely high (in thousands), and the data really does not look so incoherent. Maybe it's the model. Here's the description:
Sequential with 2 Conv2D Layers (64, 32), flattened it to feed to a Dense layer of 16 and 8, and then one final Dense layer with 1 output node with no activation function (because, regression). [Also tried to scale the image values to [0,1], but no luck.]
As a noob in this domain, I have no idea where to begin to know where I could be going wrong. If it's the way I'm loading up the data, can anyone guide me to how to go about with that. Thanks.
Just Observed:
The target test values are in the range 1 - 15 with whole numbers as the target value. The predictions are all the same, with a value around 0.00065. What could be the reason for this behaviour?
neural-network deep-learning keras regression
neural-network deep-learning keras regression
New contributor
New contributor
edited 13 hours ago
thegravity
New contributor
asked 13 hours ago
thegravitythegravity
11
11
New contributor
New contributor
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
});
}
});
thegravity is a new contributor. Be nice, and check out our Code of Conduct.
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%2f46578%2fkeras-loss-value-extremely-high-prediction-result-same%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
thegravity is a new contributor. Be nice, and check out our Code of Conduct.
thegravity is a new contributor. Be nice, and check out our Code of Conduct.
thegravity is a new contributor. Be nice, and check out our Code of Conduct.
thegravity is a new contributor. Be nice, and check out our Code of Conduct.
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%2f46578%2fkeras-loss-value-extremely-high-prediction-result-same%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