Styjun

Yandex programming contest: Alarms

Can you move on your turn, and then use the Ready Action to move again on another creature's turn?

Why does the 6502 have the BIT instruction?

What are the slash markings on Gatwick's 08R/26L?

What is the intuition behind uniform continuity?

What are the benefits of cryosleep?

Select row of data if next row contains zero

Can a helicopter mask itself from Radar?

Rotated Position of Integers

Beginner's snake game using PyGame

count number of files in directory with a certain name

Differences between “pas vrai ?”, “c’est ça ?”, “hein ?”, and “n’est-ce pas ?”

Can an old DSLR be upgraded to match modern smartphone image quality

60s (or earlier) short story where each colony has one person who doesn't connect well with others who is there for being able to absorb knowledge

In what episode of TOS did a character on the bridge make a comment about raising the number 1 to some power?

Are UK pensions taxed twice?

How can I include a header file that contains `>` in its name?

Adding strings in lists together

Why does the UK have more political parties than the US?

How do I subvert the tropes of a train heist?

Is the world in Game of Thrones spherical or flat?

What is/are this/these giant NASA box(es)?

Comment dit-on « I’ll tell you what » ?

What is game ban VS VAC ban in steam?

Predicting from the middle of a Keras model

Regularisers in Keras vs. CaffeKeras: Use the same layer in different models (share weights)Keras model gets constant loss and accuracyKeras model to predict number sequenceChange the input size in KerasMultilayer Seq2Seq model with LSTM in KerasRemove middle layers in the pre-trained VGG16 model in KerasImplementation of Adversarial Loss In Kerasreuse middle layer as input for another model in KerasCaffe to Keras conversion of grouped convolution

.everyoneloves__top-leaderboard:empty,.everyoneloves__mid-leaderboard:empty,.everyoneloves__bot-mid-leaderboard:empty height:90px;width:728px;box-sizing:border-box;

0

I am trying to develop an auto-encoder for compressing images using Keras. I was able to train it and to compress images, but I am struggling with the decoder part of it. Specifically, given a compressed image, I don't know how to use the model to de-compress it.

This is what I have:

 input_layer = keras.layers.Input(shape=(64, 64, 3))
 code_layer = build_encoder(input_layer, size_of_code) # add some convolution layers and max-pooling
 output_layer = build_decoder(code_layer) # add some convolution layers and up-sampling

 autoencoder_model = keras.models.Model(input_layer, output_layer)
 encoder_model = keras.models.Model(input_layer, code_layer)
 decoder_model = ??
 autoencoder_model.compile(optimizer='adam', loss='binary_crossentropy')

using the code above I can train the autoencoder_model and compress the images using the encoder_model, but I don't know how to construct the decoder_model, mainly because I don't know how to insert a new input to the middle of the model.

python tensorflow keras neural-network

share|improve this question

asked Mar 24 at 10:03

monte carlomonte carlo

3017

add a comment | 

0

I am trying to develop an auto-encoder for compressing images using Keras. I was able to train it and to compress images, but I am struggling with the decoder part of it. Specifically, given a compressed image, I don't know how to use the model to de-compress it.

This is what I have:

 input_layer = keras.layers.Input(shape=(64, 64, 3))
 code_layer = build_encoder(input_layer, size_of_code) # add some convolution layers and max-pooling
 output_layer = build_decoder(code_layer) # add some convolution layers and up-sampling

 autoencoder_model = keras.models.Model(input_layer, output_layer)
 encoder_model = keras.models.Model(input_layer, code_layer)
 decoder_model = ??
 autoencoder_model.compile(optimizer='adam', loss='binary_crossentropy')

using the code above I can train the autoencoder_model and compress the images using the encoder_model, but I don't know how to construct the decoder_model, mainly because I don't know how to insert a new input to the middle of the model.

python tensorflow keras neural-network

share|improve this question

asked Mar 24 at 10:03

monte carlomonte carlo

3017

add a comment | 

I am trying to develop an auto-encoder for compressing images using Keras. I was able to train it and to compress images, but I am struggling with the decoder part of it. Specifically, given a compressed image, I don't know how to use the model to de-compress it.

This is what I have:

 input_layer = keras.layers.Input(shape=(64, 64, 3))
 code_layer = build_encoder(input_layer, size_of_code) # add some convolution layers and max-pooling
 output_layer = build_decoder(code_layer) # add some convolution layers and up-sampling

 autoencoder_model = keras.models.Model(input_layer, output_layer)
 encoder_model = keras.models.Model(input_layer, code_layer)
 decoder_model = ??
 autoencoder_model.compile(optimizer='adam', loss='binary_crossentropy')

using the code above I can train the autoencoder_model and compress the images using the encoder_model, but I don't know how to construct the decoder_model, mainly because I don't know how to insert a new input to the middle of the model.

python tensorflow keras neural-network

share|improve this question

asked Mar 24 at 10:03

monte carlomonte carlo

3017

 input_layer = keras.layers.Input(shape=(64, 64, 3))
 code_layer = build_encoder(input_layer, size_of_code) # add some convolution layers and max-pooling
 output_layer = build_decoder(code_layer) # add some convolution layers and up-sampling

 autoencoder_model = keras.models.Model(input_layer, output_layer)
 encoder_model = keras.models.Model(input_layer, code_layer)
 decoder_model = ??
 autoencoder_model.compile(optimizer='adam', loss='binary_crossentropy')

share|improve this question

asked Mar 24 at 10:03

monte carlomonte carlo

3017

share|improve this question

asked Mar 24 at 10:03

monte carlomonte carlo

3017

asked Mar 24 at 10:03

monte carlomonte carlo

3017

asked Mar 24 at 10:03

monte carlomonte carlo

3017

1 Answer
1

active

oldest

votes

2

Like this. Instead of the code_layer, need to define an input layer and build the decoder model with that input.

latent_inputs = keras.layers.Input(shape=(size_of_code))
output_layer = build_decoder(latent_inputs) # add some convolution layers and up-sampling
decoder_model = keras.models.Model(latent_inputs, output_layer)

You can refer this complete VAE example:

https://github.com/keras-team/keras/blob/master/examples/variational_autoencoder.py

share|improve this answer

answered Mar 24 at 15:23

Manoj MohanManoj Mohan

1,696510

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
);



);

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%2f55322619%2fpredicting-from-the-middle-of-a-keras-model%23new-answer', 'question_page');

);

Post as a guest

Name

Email

Required, but never shown

2

Like this. Instead of the code_layer, need to define an input layer and build the decoder model with that input.

latent_inputs = keras.layers.Input(shape=(size_of_code))
output_layer = build_decoder(latent_inputs) # add some convolution layers and up-sampling
decoder_model = keras.models.Model(latent_inputs, output_layer)

You can refer this complete VAE example:

https://github.com/keras-team/keras/blob/master/examples/variational_autoencoder.py

share|improve this answer

answered Mar 24 at 15:23

Manoj MohanManoj Mohan

1,696510

add a comment | 

2

Like this. Instead of the code_layer, need to define an input layer and build the decoder model with that input.

latent_inputs = keras.layers.Input(shape=(size_of_code))
output_layer = build_decoder(latent_inputs) # add some convolution layers and up-sampling
decoder_model = keras.models.Model(latent_inputs, output_layer)

You can refer this complete VAE example:

https://github.com/keras-team/keras/blob/master/examples/variational_autoencoder.py

share|improve this answer

answered Mar 24 at 15:23

Manoj MohanManoj Mohan

1,696510

add a comment | 

Like this. Instead of the code_layer, need to define an input layer and build the decoder model with that input.

latent_inputs = keras.layers.Input(shape=(size_of_code))
output_layer = build_decoder(latent_inputs) # add some convolution layers and up-sampling
decoder_model = keras.models.Model(latent_inputs, output_layer)

You can refer this complete VAE example:

https://github.com/keras-team/keras/blob/master/examples/variational_autoencoder.py

share|improve this answer

answered Mar 24 at 15:23

Manoj MohanManoj Mohan

1,696510

latent_inputs = keras.layers.Input(shape=(size_of_code))
output_layer = build_decoder(latent_inputs) # add some convolution layers and up-sampling
decoder_model = keras.models.Model(latent_inputs, output_layer)

share|improve this answer

answered Mar 24 at 15:23

Manoj MohanManoj Mohan

1,696510

answered Mar 24 at 15:23

Manoj MohanManoj Mohan

1,696510

answered Mar 24 at 15:23

Manoj MohanManoj Mohan

1,696510