Where did Otto von Bismarck say "lying awake all night, hating"?

Who are the people reviewing far more papers than they're submitting for review?

Why don't airports use arresting gears to recover energy from landing passenger planes?

Do encumbered characters suffer any penalties when exploring and/or traveling?

All numbers twice in a 7x7 Minesweeper grid

Better way to obtain a smaller vertical line | scaled to math?

Wrong Schengen Visa exit stamp on my passport, who can I complain to?

Very lazy puppy

Tips for remembering the order of parameters for ln?

Which block header fields are miners able to change in an effort to avoid having to recalculate the Merkle Root?

Bash attempts to write two shell prompts?

What is Cousin Itt in The Addams Family?

Should the pagination be reset when changing the order?

How far away from you does grass spread?

How do rulers get rich from war?

Explanation of 申し訳ございません

Is a global DNS record a security risk for phpMyAdmin?

What does the Free Recovery sign (UK) actually mean?

Three knights searching for a princess in a castle

Why would a fighter use the afterburner and air brakes at the same time?

Inquiry answerer

Paradox regarding phase transitions in relativistic systems

What exactly is a web font, and what does converting to one involve?

Amortized Loans seem to benefit the bank more than the customer

Access Weight and Bias with nn::sequential

Set neural network initial weight values in C++ torch

.everyoneloves__top-leaderboard:empty,.everyoneloves__mid-leaderboard:empty,.everyoneloves__bot-mid-leaderboard:empty margin-bottom:0;

1

If I define std::vector<torch::nn::Linear> linear_layers; and fill this vector with some torch::nn::Linear objects, then I can access the weight and bias values by linear_layers[k].weight and linear_layers[k].bias. Same feature is available with other layer types, e.g., torch::nn::Conv2d.

If create my network using nn::sequential and then push back either of Linear or Conv2d I cannot access the weight and bias directly. Now, my question is how can I access the weight and bias values of each layer when I have used nn::sequential?

Thanks,
Afshin

libtorch

share|improve this question

asked Mar 28 at 13:21

Afshin OroojlooyAfshin Oroojlooy

55299 silver badges2525 bronze badges

add a comment
 | 

1

If I define std::vector<torch::nn::Linear> linear_layers; and fill this vector with some torch::nn::Linear objects, then I can access the weight and bias values by linear_layers[k].weight and linear_layers[k].bias. Same feature is available with other layer types, e.g., torch::nn::Conv2d.

If create my network using nn::sequential and then push back either of Linear or Conv2d I cannot access the weight and bias directly. Now, my question is how can I access the weight and bias values of each layer when I have used nn::sequential?

Thanks,
Afshin

libtorch

share|improve this question

asked Mar 28 at 13:21

Afshin OroojlooyAfshin Oroojlooy

55299 silver badges2525 bronze badges

add a comment
 | 

If I define std::vector<torch::nn::Linear> linear_layers; and fill this vector with some torch::nn::Linear objects, then I can access the weight and bias values by linear_layers[k].weight and linear_layers[k].bias. Same feature is available with other layer types, e.g., torch::nn::Conv2d.

If create my network using nn::sequential and then push back either of Linear or Conv2d I cannot access the weight and bias directly. Now, my question is how can I access the weight and bias values of each layer when I have used nn::sequential?

Thanks,
Afshin

libtorch

share|improve this question

asked Mar 28 at 13:21

Afshin OroojlooyAfshin Oroojlooy

55299 silver badges2525 bronze badges

share|improve this question

asked Mar 28 at 13:21

Afshin OroojlooyAfshin Oroojlooy

55299 silver badges2525 bronze badges

share|improve this question

asked Mar 28 at 13:21

Afshin OroojlooyAfshin Oroojlooy

55299 silver badges2525 bronze badges

asked Mar 28 at 13:21

Afshin OroojlooyAfshin Oroojlooy

55299 silver badges2525 bronze badges

asked Mar 28 at 13:21

Afshin OroojlooyAfshin Oroojlooy

55299 silver badges2525 bronze badges

1 Answer
1

active

oldest

votes

1

Here is the soultion: [see the link https://discuss.pytorch.org/t/common-class-of-linear-conv-etc/39987/8 ]

include

using namespace torch;
using namespace torch::nn;

int main()

auto net = Sequential(Conv2d(1 /input channels/, 1 /output channels/, 2 /kernel size/),
Conv2d(1, 1, 2));

for (auto& p : net->named_parameters()) 

 NoGradGuard no_grad;

 // Access name.
 std::cout << p.key() << std::endl;

 // Access weigth and bias.
 p.value().zero_(); // set all zero
 std::cout << p.value() << std::endl;


return 0;

The layers of a sequential, have the following naming convention: ., e.g. see the console output

0.weight # name of the layer
(1,1,.,.) = 
 0 0
 0 0
[ Variable[CPUFloatType]1,1,2,2 ]
0.bias
 0
[ Variable[CPUFloatType]1 ]
1.weight
(1,1,.,.) = 
 0 0
 0 0
[ Variable[CPUFloatType]1,1,2,2 ]
1.bias
 0
[ Variable[CPUFloatType]1 ]

share|improve this answer

edited Jun 27 at 18:21

Afshin Oroojlooy

55299 silver badges2525 bronze badges

answered Jun 27 at 18:19

Davood HajinezhadDavood Hajinezhad

3144 bronze badges

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/4.0/"u003ecc by-sa 4.0 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

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

Name

Email

Required, but never shown

StackExchange.ready(
function ()
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f55398723%2faccess-weight-and-bias-with-nnsequential%23new-answer', 'question_page');

);

Post as a guest

Name

Email

Required, but never shown

1

Here is the soultion: [see the link https://discuss.pytorch.org/t/common-class-of-linear-conv-etc/39987/8 ]

include

using namespace torch;
using namespace torch::nn;

int main()

auto net = Sequential(Conv2d(1 /input channels/, 1 /output channels/, 2 /kernel size/),
Conv2d(1, 1, 2));

for (auto& p : net->named_parameters()) 

 NoGradGuard no_grad;

 // Access name.
 std::cout << p.key() << std::endl;

 // Access weigth and bias.
 p.value().zero_(); // set all zero
 std::cout << p.value() << std::endl;


return 0;

The layers of a sequential, have the following naming convention: ., e.g. see the console output

0.weight # name of the layer
(1,1,.,.) = 
 0 0
 0 0
[ Variable[CPUFloatType]1,1,2,2 ]
0.bias
 0
[ Variable[CPUFloatType]1 ]
1.weight
(1,1,.,.) = 
 0 0
 0 0
[ Variable[CPUFloatType]1,1,2,2 ]
1.bias
 0
[ Variable[CPUFloatType]1 ]

share|improve this answer

edited Jun 27 at 18:21

Afshin Oroojlooy

55299 silver badges2525 bronze badges

answered Jun 27 at 18:19

Davood HajinezhadDavood Hajinezhad

3144 bronze badges

add a comment
 | 

1

Here is the soultion: [see the link https://discuss.pytorch.org/t/common-class-of-linear-conv-etc/39987/8 ]

include

using namespace torch;
using namespace torch::nn;

int main()

auto net = Sequential(Conv2d(1 /input channels/, 1 /output channels/, 2 /kernel size/),
Conv2d(1, 1, 2));

for (auto& p : net->named_parameters()) 

 NoGradGuard no_grad;

 // Access name.
 std::cout << p.key() << std::endl;

 // Access weigth and bias.
 p.value().zero_(); // set all zero
 std::cout << p.value() << std::endl;


return 0;

The layers of a sequential, have the following naming convention: ., e.g. see the console output

0.weight # name of the layer
(1,1,.,.) = 
 0 0
 0 0
[ Variable[CPUFloatType]1,1,2,2 ]
0.bias
 0
[ Variable[CPUFloatType]1 ]
1.weight
(1,1,.,.) = 
 0 0
 0 0
[ Variable[CPUFloatType]1,1,2,2 ]
1.bias
 0
[ Variable[CPUFloatType]1 ]

share|improve this answer

edited Jun 27 at 18:21

Afshin Oroojlooy

55299 silver badges2525 bronze badges

answered Jun 27 at 18:19

Davood HajinezhadDavood Hajinezhad

3144 bronze badges

add a comment
 | 

Here is the soultion: [see the link https://discuss.pytorch.org/t/common-class-of-linear-conv-etc/39987/8 ]

include

using namespace torch;
using namespace torch::nn;

int main()

auto net = Sequential(Conv2d(1 /input channels/, 1 /output channels/, 2 /kernel size/),
Conv2d(1, 1, 2));

for (auto& p : net->named_parameters()) 

 NoGradGuard no_grad;

 // Access name.
 std::cout << p.key() << std::endl;

 // Access weigth and bias.
 p.value().zero_(); // set all zero
 std::cout << p.value() << std::endl;


return 0;

The layers of a sequential, have the following naming convention: ., e.g. see the console output

0.weight # name of the layer
(1,1,.,.) = 
 0 0
 0 0
[ Variable[CPUFloatType]1,1,2,2 ]
0.bias
 0
[ Variable[CPUFloatType]1 ]
1.weight
(1,1,.,.) = 
 0 0
 0 0
[ Variable[CPUFloatType]1,1,2,2 ]
1.bias
 0
[ Variable[CPUFloatType]1 ]

share|improve this answer

edited Jun 27 at 18:21

Afshin Oroojlooy

55299 silver badges2525 bronze badges

answered Jun 27 at 18:19

Davood HajinezhadDavood Hajinezhad

3144 bronze badges

for (auto& p : net->named_parameters()) 

 NoGradGuard no_grad;

 // Access name.
 std::cout << p.key() << std::endl;

 // Access weigth and bias.
 p.value().zero_(); // set all zero
 std::cout << p.value() << std::endl;


return 0;

0.weight # name of the layer
(1,1,.,.) = 
 0 0
 0 0
[ Variable[CPUFloatType]1,1,2,2 ]
0.bias
 0
[ Variable[CPUFloatType]1 ]
1.weight
(1,1,.,.) = 
 0 0
 0 0
[ Variable[CPUFloatType]1,1,2,2 ]
1.bias
 0
[ Variable[CPUFloatType]1 ]

share|improve this answer

edited Jun 27 at 18:21

Afshin Oroojlooy

55299 silver badges2525 bronze badges

answered Jun 27 at 18:19

Davood HajinezhadDavood Hajinezhad

3144 bronze badges

edited Jun 27 at 18:21

Afshin Oroojlooy

55299 silver badges2525 bronze badges

edited Jun 27 at 18:21

Afshin Oroojlooy

55299 silver badges2525 bronze badges

answered Jun 27 at 18:19

Davood HajinezhadDavood Hajinezhad

3144 bronze badges

answered Jun 27 at 18:19

Davood HajinezhadDavood Hajinezhad

3144 bronze badges