Styjun

How dangerous are my worn rims?

Single tx included in two different blocks

Generating numbers with cubes

Job interview by video at home and privacy concerns

The answer is a girl's name (my future granddaughter) - can anyone help?

Why did they use ultrafast diodes in a 50 or 60 Hz bridge?

French license plates

Why has Speaker Pelosi been so hesitant to impeach President Trump?

Is there a way to make an animal companion able to read a language?

Giving a good fancy look to a simple table

What does a textbook look like while you are writing it?

Avoiding dust scattering when you drill

How to level a picture frame hung on a single nail?

Ĉi tie or ĉi-tie? Why do people sometimes hyphenate ĉi tie?

Rank-one positive decomposition for a entry-wise positive positive definite matrix

Do jackscrews suffer from blowdown?

Citing CPLEX 12.9

Everyone Gets a Window Seat

IEEE 754 square root with Newton-Raphson

Would a horse be sufficient buffer to prevent injury when falling from a great height?

Parent asking for money after I moved out

Bothered by watching coworkers slacking off

Re-entering the UK after overstaying in 2008

Could Boris Johnson face criminal charges for illegally proroguing Parliament?

How to export/save a text classifier after fine-tuned on a pre-trained BERT

How to train a set of images (apart form the tutorial) and use it to classify in Tensorflow?How to export a trained model from tensorflow and how correctly to import in opencv dnn?Export Tensorflow EstimatorHow to feed input into one layer in a tensorflow pre-trained model?TypeError: __init__() missing 2 required positional arguments: 'message' and 'code'how to solve the no 'estimator' in tensorflow?Converting Tensorflow Graph to use Estimator, get 'TypeError: data type not understood' at loss function using `sampled_softmax_loss` or `nce_loss`Tensorflow Estimator causes 'data type not understood' error at `sampled_softmax_loss` or `nce_loss` for Tensorflow's official word2vec implementationTensor conversion requested dtype int32 for Tensor with dtype int64 - while estimator.export_savedmodelSaving a custom tf.estimator trained model for tensorflow serving

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

5

I followed the steps on the Predicting Movie Reviews with BERT on TF Hub here: https://colab.research.google.com/github/google-research/bert/blob/master/predicting_movie_reviews_with_bert_on_tf_hub.ipynb#scrollTo=PPVEXhNjYXC-.

At the end, how do I export the model to be used/loaded as a classifier later?

I found a link (https://guillaumegenthial.github.io/serving-tensorflow-estimator.html) that shows that I can export the estimator as a tf.saved_model. However, I got stuck on creating the 'serving_input_receiver_fn()'.

python tensorflow tensorflow-estimator

share|improve this question

asked Mar 28 at 21:16

ebaikebaik

2633 bronze badges

• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  What about using pickle to serialize it?
  
  – Alberto Bonsanto
  Mar 28 at 21:19
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  @AlbertoBonsanto Thanks for your input. Could you elaborate on that? Do you mean dump the 'estimator' using pickle?
  
  – ebaik
  Mar 29 at 15:55
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  any luck on this? i am stuck in same situation?
  
  – Upasana Mittal
  Jul 28 at 4:23
  

  
  
  
  

add a comment
 | 

5

I followed the steps on the Predicting Movie Reviews with BERT on TF Hub here: https://colab.research.google.com/github/google-research/bert/blob/master/predicting_movie_reviews_with_bert_on_tf_hub.ipynb#scrollTo=PPVEXhNjYXC-.

At the end, how do I export the model to be used/loaded as a classifier later?

I found a link (https://guillaumegenthial.github.io/serving-tensorflow-estimator.html) that shows that I can export the estimator as a tf.saved_model. However, I got stuck on creating the 'serving_input_receiver_fn()'.

python tensorflow tensorflow-estimator

share|improve this question

asked Mar 28 at 21:16

ebaikebaik

2633 bronze badges

• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  What about using pickle to serialize it?
  
  – Alberto Bonsanto
  Mar 28 at 21:19
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  @AlbertoBonsanto Thanks for your input. Could you elaborate on that? Do you mean dump the 'estimator' using pickle?
  
  – ebaik
  Mar 29 at 15:55
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  any luck on this? i am stuck in same situation?
  
  – Upasana Mittal
  Jul 28 at 4:23
  

  
  
  
  

add a comment
 | 

I followed the steps on the Predicting Movie Reviews with BERT on TF Hub here: https://colab.research.google.com/github/google-research/bert/blob/master/predicting_movie_reviews_with_bert_on_tf_hub.ipynb#scrollTo=PPVEXhNjYXC-.

At the end, how do I export the model to be used/loaded as a classifier later?

I found a link (https://guillaumegenthial.github.io/serving-tensorflow-estimator.html) that shows that I can export the estimator as a tf.saved_model. However, I got stuck on creating the 'serving_input_receiver_fn()'.

python tensorflow tensorflow-estimator

share|improve this question

asked Mar 28 at 21:16

ebaikebaik

2633 bronze badges

share|improve this question

asked Mar 28 at 21:16

ebaikebaik

2633 bronze badges

share|improve this question

asked Mar 28 at 21:16

ebaikebaik

2633 bronze badges

asked Mar 28 at 21:16

ebaikebaik

2633 bronze badges

asked Mar 28 at 21:16

ebaikebaik

2633 bronze badges