Styjun

Am I allowed to use personal conversation as a source?

Why force the nose of 737 Max down in the first place?

How can I say in Russian "they cannot make the tournament attractive by itself"?

How can I rectify up to 85 kV

How many oliphaunts died in all of the Lord of the Rings battles?

What is the most efficient way to write 'for' loops in Matlab?

Isolated audio without a transformer

What is the difference between 1/3, 1/2, and full casters?

Pointwise convergence of uniformly continuous functions to zero, but not uniformly

Why is the number of local variables used in a Java bytecode method not the most economical?

Please explain joy and/or the Kimatthiyasutta

Melee or Ranged attacks by Monsters, no distinction in modifiers?

If Trump gets impeached, how long would Pence be president?

What is this spacecraft tethered to another spacecraft in LEO (vintage)

Is there an antonym for "spicy" or "hot" regarding food (NOT "seasoned" but "spicy")?

Did the IBM PC use the 8088's NMI line?

What is the use of で in this sentence?

Why can't my huge trees be chopped down?

Unethical behavior : should I report it?

Converting 8V AC to 8V DC - bridge rectifier gets very hot while idling

Character is called by their first initial. How do I write it?

If my pay period is split between 2 calendar years, which tax year do I file them in?

How do I stop my characters falling in love?

How to tar a list of directories only if they exist

Merge Keras Embeddings with normal models into one sequential Model

Merge 2 sequential models in Kerasloss, val_loss, acc and val_acc do not update at all over epochsUserWarning: Update your `Dense` call to the Keras 2 API:Implement perceptual loss with pretrained VGG using kerasHow does Keras read input data?Building an LSTM net with an embedding layer in KerasGet increasing loss and poor performance on Variational AutoencoderHow to save and reuse all settings for a keras model?scipy.ndimage.zoom is taking long time even on small arraysEpoch's steps taking too long on GPU'Sequential' object has no attribute 'loss' - When I used GridSearchCV to tuning my Keras model

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

0

I want to merge keras embeddings with normal models into one sequential model as in the notebook in this repository

from keras.layers import *
from keras.models import *

models = []

for categoical_var in categorical_vars :
 model = Sequential()
 model.reset_states( )
 no_of_unique_cat = df[categoical_var].nunique()
 embedding_size = min(np.ceil((no_of_unique_cat)/2), 50 )
 embedding_size = int(embedding_size)
 model.add( Embedding( no_of_unique_cat+1, embedding_size, input_length = 1 ) )
 model.add(Reshape(target_shape=(embedding_size,)))
 models.append( model )


model_rest = Sequential()
model_rest.add(Dense( 64 , input_dim = 6 ))
model_rest.reset_states( )
models.append(model_rest)

full_model = Sequential()
full_model.add(Merge(models, mode='concat'))

full_model.add(Dense(512))
full_model.add(Activation('sigmoid'))
full_model.add(Dropout(0.2))

full_model.add(Dense(32))
full_model.add(Activation('sigmoid'))
full_model.add(Dropout(0.2))

full_model.add(Dense(1))

full_model.compile(loss='mean_squared_error', optimizer='Adam',metrics=['mse','mape'])

The problem is new versions of keras doesn't use 'Merge' any more, I tried to use Concatenate, but it does not work with sequential models

Any help? I looked in to this, but it is a different problem

tensorflow keras neural-network deep-learning keras-layer

share|improve this question

asked Mar 26 at 18:48

owise owise

46655 silver badges1919 bronze badges

• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  You should implement this using the Functional API, forget about the Sequential API.
  
  – Matias Valdenegro
  Mar 26 at 21:18
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  Keras still has Merge layers. Even if it does not, you can easily implement it through the Lambda layer
  
  – pitfall
  Mar 27 at 5:49
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  Can you please elaborate on this?
  
  –  owise
  Mar 27 at 14:38
  

  
  
  
  

add a comment | 

0

I want to merge keras embeddings with normal models into one sequential model as in the notebook in this repository

from keras.layers import *
from keras.models import *

models = []

for categoical_var in categorical_vars :
 model = Sequential()
 model.reset_states( )
 no_of_unique_cat = df[categoical_var].nunique()
 embedding_size = min(np.ceil((no_of_unique_cat)/2), 50 )
 embedding_size = int(embedding_size)
 model.add( Embedding( no_of_unique_cat+1, embedding_size, input_length = 1 ) )
 model.add(Reshape(target_shape=(embedding_size,)))
 models.append( model )


model_rest = Sequential()
model_rest.add(Dense( 64 , input_dim = 6 ))
model_rest.reset_states( )
models.append(model_rest)

full_model = Sequential()
full_model.add(Merge(models, mode='concat'))

full_model.add(Dense(512))
full_model.add(Activation('sigmoid'))
full_model.add(Dropout(0.2))

full_model.add(Dense(32))
full_model.add(Activation('sigmoid'))
full_model.add(Dropout(0.2))

full_model.add(Dense(1))

full_model.compile(loss='mean_squared_error', optimizer='Adam',metrics=['mse','mape'])

The problem is new versions of keras doesn't use 'Merge' any more, I tried to use Concatenate, but it does not work with sequential models

Any help? I looked in to this, but it is a different problem

tensorflow keras neural-network deep-learning keras-layer

share|improve this question

asked Mar 26 at 18:48

owise owise

46655 silver badges1919 bronze badges

• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  You should implement this using the Functional API, forget about the Sequential API.
  
  – Matias Valdenegro
  Mar 26 at 21:18
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  Keras still has Merge layers. Even if it does not, you can easily implement it through the Lambda layer
  
  – pitfall
  Mar 27 at 5:49
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  Can you please elaborate on this?
  
  –  owise
  Mar 27 at 14:38
  

  
  
  
  

add a comment | 

I want to merge keras embeddings with normal models into one sequential model as in the notebook in this repository

from keras.layers import *
from keras.models import *

models = []

for categoical_var in categorical_vars :
 model = Sequential()
 model.reset_states( )
 no_of_unique_cat = df[categoical_var].nunique()
 embedding_size = min(np.ceil((no_of_unique_cat)/2), 50 )
 embedding_size = int(embedding_size)
 model.add( Embedding( no_of_unique_cat+1, embedding_size, input_length = 1 ) )
 model.add(Reshape(target_shape=(embedding_size,)))
 models.append( model )


model_rest = Sequential()
model_rest.add(Dense( 64 , input_dim = 6 ))
model_rest.reset_states( )
models.append(model_rest)

full_model = Sequential()
full_model.add(Merge(models, mode='concat'))

full_model.add(Dense(512))
full_model.add(Activation('sigmoid'))
full_model.add(Dropout(0.2))

full_model.add(Dense(32))
full_model.add(Activation('sigmoid'))
full_model.add(Dropout(0.2))

full_model.add(Dense(1))

full_model.compile(loss='mean_squared_error', optimizer='Adam',metrics=['mse','mape'])

The problem is new versions of keras doesn't use 'Merge' any more, I tried to use Concatenate, but it does not work with sequential models

Any help? I looked in to this, but it is a different problem

tensorflow keras neural-network deep-learning keras-layer

share|improve this question

asked Mar 26 at 18:48

owise owise

46655 silver badges1919 bronze badges

from keras.layers import *
from keras.models import *

models = []

for categoical_var in categorical_vars :
 model = Sequential()
 model.reset_states( )
 no_of_unique_cat = df[categoical_var].nunique()
 embedding_size = min(np.ceil((no_of_unique_cat)/2), 50 )
 embedding_size = int(embedding_size)
 model.add( Embedding( no_of_unique_cat+1, embedding_size, input_length = 1 ) )
 model.add(Reshape(target_shape=(embedding_size,)))
 models.append( model )


model_rest = Sequential()
model_rest.add(Dense( 64 , input_dim = 6 ))
model_rest.reset_states( )
models.append(model_rest)

full_model = Sequential()
full_model.add(Merge(models, mode='concat'))

full_model.add(Dense(512))
full_model.add(Activation('sigmoid'))
full_model.add(Dropout(0.2))

full_model.add(Dense(32))
full_model.add(Activation('sigmoid'))
full_model.add(Dropout(0.2))

full_model.add(Dense(1))

full_model.compile(loss='mean_squared_error', optimizer='Adam',metrics=['mse','mape'])

share|improve this question

asked Mar 26 at 18:48

owise owise

46655 silver badges1919 bronze badges

share|improve this question

asked Mar 26 at 18:48

owise owise

46655 silver badges1919 bronze badges

asked Mar 26 at 18:48

owise owise

46655 silver badges1919 bronze badges

asked Mar 26 at 18:48

owise owise

46655 silver badges1919 bronze badges