Styjun

Is it safe to use c_str() on a temporary string?

What can we do to stop prior company from asking us questions?

% symbol leads to superlong (forever?) compilations

How do spells that require an ability check vs. the caster's spell save DC work?

Return the Closest Prime Number

How do I get the green key off the shelf in the Dobby level of Lego Harry Potter 2?

Why didn't Theresa May consult with Parliament before negotiating a deal with the EU?

Only print output after finding pattern

Example of a Mathematician/Physicist whose Other Publications during their PhD eclipsed their PhD Thesis

What is the purpose of the Evocation wizard's Potent Cantrip feature?

WOW air has ceased operation, can I get my tickets refunded?

How long to clear the 'suck zone' of a turbofan after start is initiated?

Does the Brexit deal have to be agreed by both Houses?

Would this house-rule that treats advantage as a +1 to the roll instead (and disadvantage as -1) and allows them to stack be balanced?

Why here is plural "We went to the movies last night."

Term for the "extreme-extension" version of a straw man fallacy?

Is the concept of a "numerable" fiber bundle really useful or an empty generalization?

Text adventure game code

What does this shorthand mean?

Why do remote companies require working in the US?

Describing a person. What needs to be mentioned?

Whats the best way to handle refactoring a big file?

Unreliable Magic - Is it worth it?

How do scammers retract money, while you can’t?

Using keras.layers.Add() in a keras.sequential model

1

Using TF 2.0 and tfp probability layers, I have constructed a keras.sequential model. I would like to export it for serving with TensorFlow Serving, and I would like to include the preprocessing and post processing steps in the servable.

My preprocessing steps are fairly simple-- fill NAs with explicit values, encoding a few strings as floats, normalize inputs, and denormalize outputs. For training, I have been doing the pre/post processing with pandas and numpy.

I know that I can export my Keras model's weights, wrap the keras.sequential model's architecture in a bigger TensorFlow graph, use low-level ops like tf.math.subtract(inputs, vector_of_feature_means) to do pre/post processing operations, define tf.placeholders for my inputs and outputs, and make a servable, but I feel like there has to be a cleaner way of doing this.

Is it possible to use keras.layers.Add() and keras.layers.Multiply() in a keras.sequence model for explicit preprocessing steps, or is there some more standard way of doing these things?

tensorflow machine-learning keras tensorflow-serving tf.keras

share|improve this question

edited Mar 22 at 0:07

alift

176116

asked Mar 21 at 16:14

James McKeownJames McKeown

1147

• 
  
  
  

  
  1
  

  
  
  
  
  
  

  
  
  stackoverflow.com/questions/41672114/… is related, but a bit outdated
  
  – James McKeown
  Mar 21 at 16:48
  

  
  
  
  

add a comment | 

1

Using TF 2.0 and tfp probability layers, I have constructed a keras.sequential model. I would like to export it for serving with TensorFlow Serving, and I would like to include the preprocessing and post processing steps in the servable.

My preprocessing steps are fairly simple-- fill NAs with explicit values, encoding a few strings as floats, normalize inputs, and denormalize outputs. For training, I have been doing the pre/post processing with pandas and numpy.

I know that I can export my Keras model's weights, wrap the keras.sequential model's architecture in a bigger TensorFlow graph, use low-level ops like tf.math.subtract(inputs, vector_of_feature_means) to do pre/post processing operations, define tf.placeholders for my inputs and outputs, and make a servable, but I feel like there has to be a cleaner way of doing this.

Is it possible to use keras.layers.Add() and keras.layers.Multiply() in a keras.sequence model for explicit preprocessing steps, or is there some more standard way of doing these things?

tensorflow machine-learning keras tensorflow-serving tf.keras

share|improve this question

edited Mar 22 at 0:07

alift

176116

asked Mar 21 at 16:14

James McKeownJames McKeown

1147

• 
  
  
  

  
  1
  

  
  
  
  
  
  

  
  
  stackoverflow.com/questions/41672114/… is related, but a bit outdated
  
  – James McKeown
  Mar 21 at 16:48
  

  
  
  
  

add a comment | 

Using TF 2.0 and tfp probability layers, I have constructed a keras.sequential model. I would like to export it for serving with TensorFlow Serving, and I would like to include the preprocessing and post processing steps in the servable.

My preprocessing steps are fairly simple-- fill NAs with explicit values, encoding a few strings as floats, normalize inputs, and denormalize outputs. For training, I have been doing the pre/post processing with pandas and numpy.

I know that I can export my Keras model's weights, wrap the keras.sequential model's architecture in a bigger TensorFlow graph, use low-level ops like tf.math.subtract(inputs, vector_of_feature_means) to do pre/post processing operations, define tf.placeholders for my inputs and outputs, and make a servable, but I feel like there has to be a cleaner way of doing this.

Is it possible to use keras.layers.Add() and keras.layers.Multiply() in a keras.sequence model for explicit preprocessing steps, or is there some more standard way of doing these things?

tensorflow machine-learning keras tensorflow-serving tf.keras

share|improve this question

edited Mar 22 at 0:07

alift

176116

asked Mar 21 at 16:14

James McKeownJames McKeown

1147

share|improve this question

edited Mar 22 at 0:07

alift

176116

asked Mar 21 at 16:14

James McKeownJames McKeown

1147

share|improve this question

edited Mar 22 at 0:07

alift

176116

asked Mar 21 at 16:14

James McKeownJames McKeown

1147

edited Mar 22 at 0:07

alift

176116

edited Mar 22 at 0:07

alift

176116

asked Mar 21 at 16:14

James McKeownJames McKeown

1147

asked Mar 21 at 16:14

James McKeownJames McKeown

1147