Learnable parameters in DNN The Next CEO of Stack Overflow2019 Community Moderator ElectionStudying machine learning algorithms: depth of understanding vs. number of algorithmsWhat is conjugate gradient descent?Public cloud GPU support for TensorFlowClipping threshold of softmax layerPredict a tree structure out of nodes with different featuresWhy CNN doesn't give higher accuracy over simple MLP network? [From Keras examples]Do numerical inaccuracies play any role in training neural networks?Keras intuition/guidelines for setting epochs and batch sizeDo we need to add the sigmoid derivative term in the final layer's error value?Issues with training SSD on own dataset
What was Carter Burkes job for "the company" in "Aliens"?
Is French Guiana a (hard) EU border?
Does Germany produce more waste than the US?
Won the lottery - how do I keep the money?
How to properly draw diagonal line while using multicolumn inside tabular environment?
What connection does MS Office have to Netscape Navigator?
When "be it" is at the beginning of a sentence, what kind of structure do you call it?
What flight has the highest ratio of timezone difference to flight time?
Pulling the principal components out of a DimensionReducerFunction?
What difference does it make using sed with/without whitespaces?
How to count occurrences of text in a file?
what's the use of '% to gdp' type of variables?
Is there a difference between "Fahrstuhl" and "Aufzug"?
Can I board the first leg of the flight without having final country's visa?
What day is it again?
Calculate the Mean mean of two numbers
Deriving the equation for variance
Would a grinding machine be a simple and workable propulsion system for an interplanetary spacecraft?
What steps are necessary to read a Modern SSD in Medieval Europe?
Audio Conversion With ADS1243
If Nick Fury and Coulson already knew about aliens (Kree and Skrull) why did they wait until Thor's appearance to start making weapons?
New carbon wheel brake pads after use on aluminum wheel?
Towers in the ocean; How deep can they be built?
IC has pull-down resistors on SMBus lines?
Learnable parameters in DNN
The Next CEO of Stack Overflow2019 Community Moderator ElectionStudying machine learning algorithms: depth of understanding vs. number of algorithmsWhat is conjugate gradient descent?Public cloud GPU support for TensorFlowClipping threshold of softmax layerPredict a tree structure out of nodes with different featuresWhy CNN doesn't give higher accuracy over simple MLP network? [From Keras examples]Do numerical inaccuracies play any role in training neural networks?Keras intuition/guidelines for setting epochs and batch sizeDo we need to add the sigmoid derivative term in the final layer's error value?Issues with training SSD on own dataset
$begingroup$
I've came across the term "learnable parameters" recently, and googling didn't help much as most search was describing learnable parameters in a CNN instead of a DNN. Is there any difference between the two?
How would i compute the number of learnable parameters in a DNN? Could anyone please explain what those are with an example? I'm new to machine learning so i would appreciate some help on this.
machine-learning neural-network
asked Mar 23 at 11:08
MaxxxMaxxx
1253
$endgroup$
add a comment |
$begingroup$
I've came across the term "learnable parameters" recently, and googling didn't help much as most search was describing learnable parameters in a CNN instead of a DNN. Is there any difference between the two?
How would i compute the number of learnable parameters in a DNN? Could anyone please explain what those are with an example? I'm new to machine learning so i would appreciate some help on this.
machine-learning neural-network
asked Mar 23 at 11:08
MaxxxMaxxx
1253
$endgroup$
add a comment |
$begingroup$
I've came across the term "learnable parameters" recently, and googling didn't help much as most search was describing learnable parameters in a CNN instead of a DNN. Is there any difference between the two?
How would i compute the number of learnable parameters in a DNN? Could anyone please explain what those are with an example? I'm new to machine learning so i would appreciate some help on this.
machine-learning neural-network
asked Mar 23 at 11:08
MaxxxMaxxx
1253
$endgroup$
I've came across the term "learnable parameters" recently, and googling didn't help much as most search was describing learnable parameters in a CNN instead of a DNN. Is there any difference between the two?
How would i compute the number of learnable parameters in a DNN? Could anyone please explain what those are with an example? I'm new to machine learning so i would appreciate some help on this.
machine-learning neural-network
machine-learning neural-network
asked Mar 23 at 11:08
MaxxxMaxxx
1253
asked Mar 23 at 11:08
MaxxxMaxxx
1253
asked Mar 23 at 11:08
MaxxxMaxxx
1253
asked Mar 23 at 11:08
MaxxxMaxxx
1253
asked Mar 23 at 11:08
MaxxxMaxxx
1253
1253
add a comment |
add a comment |
2 Answers
2
active
oldest
votes
$begingroup$
In neural networks in general, and in deep learning algorithms (CNN, DNN, ...) that are also based on neural networks, learnable parameters are parameters that will be learned by the model during the training procedure such weights and biases.
you can generate learnable parameters for each layer of your model.
DNN is a deep neural network, what I understand that when a neural network becomes deep, it can be said that its a deep learning model.
so, to compute a DNN or CNN learnable parameters, you can build your model using keras, keras can automatically generate the models learnables parameter for each layer, when you complete building your model type model.summary() it generates a list of learnable parameters when you run it.
this link builds a DNN model using using keras link
answered Mar 23 at 11:29
honar.cshonar.cs
18313
$endgroup$
add a comment |
$begingroup$
Imagine a simple input vector $boldsymbolx=[x_1,x_2,x_3]^T$, with $x_iin 0,1,ldots,255$. If our neural network has a layer which normalizes the inputs into the range of rational numbers between $0$ and $1$. The output of this operation would be
$$boldsymbolx_textNorm=[1/255, 1/255, 1/255]boldsymbolx.$$
As you can see we have $3$ parameters but all are not trainable.
A more common situation is given when you do transfer learning with pretrained neural networks. In this situation, you take a trained neural network and modify it (e.g. add additional layers or chop of some layers and add your own layers). Then when you want to train the new network your data set you can decide if you want to use the pretrained parameters (non-trainable parameters) or train the network from the scratch. In the last case, you will have a model in which all parameters are trainable.
answered Mar 23 at 11:30
MachineLearnerMachineLearner
36910
$endgroup$
add a comment |
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%2f47839%2flearnable-parameters-in-dnn%23new-answer', 'question_page');
);
Post as a guest
Required, but never shown
2 Answers
2
active
oldest
votes
2 Answers
2
active
oldest
votes
active
oldest
votes
active
oldest
votes
$begingroup$
In neural networks in general, and in deep learning algorithms (CNN, DNN, ...) that are also based on neural networks, learnable parameters are parameters that will be learned by the model during the training procedure such weights and biases.
you can generate learnable parameters for each layer of your model.
DNN is a deep neural network, what I understand that when a neural network becomes deep, it can be said that its a deep learning model.
so, to compute a DNN or CNN learnable parameters, you can build your model using keras, keras can automatically generate the models learnables parameter for each layer, when you complete building your model type model.summary() it generates a list of learnable parameters when you run it.
this link builds a DNN model using using keras link
answered Mar 23 at 11:29
honar.cshonar.cs
18313
$endgroup$
add a comment |
$begingroup$
In neural networks in general, and in deep learning algorithms (CNN, DNN, ...) that are also based on neural networks, learnable parameters are parameters that will be learned by the model during the training procedure such weights and biases.
you can generate learnable parameters for each layer of your model.
DNN is a deep neural network, what I understand that when a neural network becomes deep, it can be said that its a deep learning model.
so, to compute a DNN or CNN learnable parameters, you can build your model using keras, keras can automatically generate the models learnables parameter for each layer, when you complete building your model type model.summary() it generates a list of learnable parameters when you run it.
this link builds a DNN model using using keras link
answered Mar 23 at 11:29
honar.cshonar.cs
18313
$endgroup$
add a comment |
$begingroup$
In neural networks in general, and in deep learning algorithms (CNN, DNN, ...) that are also based on neural networks, learnable parameters are parameters that will be learned by the model during the training procedure such weights and biases.
you can generate learnable parameters for each layer of your model.
DNN is a deep neural network, what I understand that when a neural network becomes deep, it can be said that its a deep learning model.
so, to compute a DNN or CNN learnable parameters, you can build your model using keras, keras can automatically generate the models learnables parameter for each layer, when you complete building your model type model.summary() it generates a list of learnable parameters when you run it.
this link builds a DNN model using using keras link
answered Mar 23 at 11:29
honar.cshonar.cs
18313
$endgroup$
In neural networks in general, and in deep learning algorithms (CNN, DNN, ...) that are also based on neural networks, learnable parameters are parameters that will be learned by the model during the training procedure such weights and biases.
you can generate learnable parameters for each layer of your model.
DNN is a deep neural network, what I understand that when a neural network becomes deep, it can be said that its a deep learning model.
so, to compute a DNN or CNN learnable parameters, you can build your model using keras, keras can automatically generate the models learnables parameter for each layer, when you complete building your model type model.summary() it generates a list of learnable parameters when you run it.
this link builds a DNN model using using keras link
answered Mar 23 at 11:29
honar.cshonar.cs
18313
answered Mar 23 at 11:29
honar.cshonar.cs
18313
answered Mar 23 at 11:29
honar.cshonar.cs
18313
answered Mar 23 at 11:29
honar.cshonar.cs
18313
18313
add a comment |
add a comment |
$begingroup$
Imagine a simple input vector $boldsymbolx=[x_1,x_2,x_3]^T$, with $x_iin 0,1,ldots,255$. If our neural network has a layer which normalizes the inputs into the range of rational numbers between $0$ and $1$. The output of this operation would be
$$boldsymbolx_textNorm=[1/255, 1/255, 1/255]boldsymbolx.$$
As you can see we have $3$ parameters but all are not trainable.
A more common situation is given when you do transfer learning with pretrained neural networks. In this situation, you take a trained neural network and modify it (e.g. add additional layers or chop of some layers and add your own layers). Then when you want to train the new network your data set you can decide if you want to use the pretrained parameters (non-trainable parameters) or train the network from the scratch. In the last case, you will have a model in which all parameters are trainable.
answered Mar 23 at 11:30
MachineLearnerMachineLearner
36910
$endgroup$
add a comment |
$begingroup$
Imagine a simple input vector $boldsymbolx=[x_1,x_2,x_3]^T$, with $x_iin 0,1,ldots,255$. If our neural network has a layer which normalizes the inputs into the range of rational numbers between $0$ and $1$. The output of this operation would be
$$boldsymbolx_textNorm=[1/255, 1/255, 1/255]boldsymbolx.$$
As you can see we have $3$ parameters but all are not trainable.
A more common situation is given when you do transfer learning with pretrained neural networks. In this situation, you take a trained neural network and modify it (e.g. add additional layers or chop of some layers and add your own layers). Then when you want to train the new network your data set you can decide if you want to use the pretrained parameters (non-trainable parameters) or train the network from the scratch. In the last case, you will have a model in which all parameters are trainable.
answered Mar 23 at 11:30
MachineLearnerMachineLearner
36910
$endgroup$
add a comment |
$begingroup$
Imagine a simple input vector $boldsymbolx=[x_1,x_2,x_3]^T$, with $x_iin 0,1,ldots,255$. If our neural network has a layer which normalizes the inputs into the range of rational numbers between $0$ and $1$. The output of this operation would be
$$boldsymbolx_textNorm=[1/255, 1/255, 1/255]boldsymbolx.$$
As you can see we have $3$ parameters but all are not trainable.
A more common situation is given when you do transfer learning with pretrained neural networks. In this situation, you take a trained neural network and modify it (e.g. add additional layers or chop of some layers and add your own layers). Then when you want to train the new network your data set you can decide if you want to use the pretrained parameters (non-trainable parameters) or train the network from the scratch. In the last case, you will have a model in which all parameters are trainable.
answered Mar 23 at 11:30
MachineLearnerMachineLearner
36910
$endgroup$
Imagine a simple input vector $boldsymbolx=[x_1,x_2,x_3]^T$, with $x_iin 0,1,ldots,255$. If our neural network has a layer which normalizes the inputs into the range of rational numbers between $0$ and $1$. The output of this operation would be
$$boldsymbolx_textNorm=[1/255, 1/255, 1/255]boldsymbolx.$$
As you can see we have $3$ parameters but all are not trainable.
A more common situation is given when you do transfer learning with pretrained neural networks. In this situation, you take a trained neural network and modify it (e.g. add additional layers or chop of some layers and add your own layers). Then when you want to train the new network your data set you can decide if you want to use the pretrained parameters (non-trainable parameters) or train the network from the scratch. In the last case, you will have a model in which all parameters are trainable.
answered Mar 23 at 11:30
MachineLearnerMachineLearner
36910
answered Mar 23 at 11:30
MachineLearnerMachineLearner
36910
answered Mar 23 at 11:30
MachineLearnerMachineLearner
36910
answered Mar 23 at 11:30
MachineLearnerMachineLearner
36910
36910
add a comment |
add a comment |
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%2f47839%2flearnable-parameters-in-dnn%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
