Styjun

Does POSIX guarantee the paths to any standard utilities?

Count rook moves 1D

How many days for hunting?

How to generate all 3×3 matrices with a,a,a,a,b,b,b,c,c？

How does speed affect lift?

Main differences between 5th edition Druid and 3.5 edition Druid

Tiny image scraper for xkcd.com

Can there be plants on the dark side of a tidally locked world?

co-son-in-law or co-brother

Do I need to get a noble in order to win Splendor?

What drugs were used in England during the High Middle Ages?

How could it be that the capo isn't changing the pitch?

If p-value is exactly 1 (1.0000000), what are the confidence interval limits?

What is the significance of 104% for throttle power and rotor speed?

What did Boris Johnson mean when he said "extra 34 billion going into the NHS"

Baking with Whey Protein

Adding transparency to ink drawing

Were the women of Travancore, India, taxed for covering their breasts by breast size?

Powering an offset stacked array of pistons

Deleting millions of records on SQL Server 14.0

What would a biological creature need in order to see the future?

Did Alan Turing's student Robin Gandy assert that Charles Babbage had no notion of a universal computing machine?

std::tuple sizeof, is it a missed optimization?

Why do many programmers abstain from using global variables?

How can I export a model with multiple data partition

Relationship between tensorflow saver, exporter and save modelhow to use SavedModelBuilder to export a nlp modelLoading trained Tensorflow model into estimatorDistributed Tensorflow Estimator execution does not trigger evaluation or exportTensorflow warm_start even if checkpoint foundHow to make optimistic restore work with TPU Estimator in Tensorflow?Serving retrained Tensorflow Hub module with new features`get_variable()` doesn't recognize existing variables for tf.estimatorfreeze_graph MemoryError for .pbtxt from TF.EstimatorAdd labels to estimator.export_saved_model when exporting a keras model for google cloud

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

0

I want to export the model from checkpoint with estimator, and the variable data is stored in one file. I want the variable data to be stored in multi files so that I can distcp it faster.

I tried launched multi workers to export, but it does not work.

Here is some of my codes:

 if self.job_name == "worker":
 if FLAGS.ckpt != '' and int(FLAGS.ckpt) > 0:
 estimator.export_savedmodel(self.export_model_path,
 receiver_fn_raw(self.params['dense_num'], self.params['feature_conf'],
 self.params['list_num']),
 checkpoint_path=self.ckpt_dir)
 else:
 estimator.export_savedmodel(self.export_model_path,
 receiver_fn_raw(self.params['dense_num'], self.params['feature_conf'],
 self.params['list_num']))

tensorflow tensorflow-serving tensorflow-estimator

share|improve this question

asked Mar 28 at 2:47

Xiang AplysiaXiang Aplysia

1255 bronze badges

add a comment | 

0

I want to export the model from checkpoint with estimator, and the variable data is stored in one file. I want the variable data to be stored in multi files so that I can distcp it faster.

I tried launched multi workers to export, but it does not work.

Here is some of my codes:

 if self.job_name == "worker":
 if FLAGS.ckpt != '' and int(FLAGS.ckpt) > 0:
 estimator.export_savedmodel(self.export_model_path,
 receiver_fn_raw(self.params['dense_num'], self.params['feature_conf'],
 self.params['list_num']),
 checkpoint_path=self.ckpt_dir)
 else:
 estimator.export_savedmodel(self.export_model_path,
 receiver_fn_raw(self.params['dense_num'], self.params['feature_conf'],
 self.params['list_num']))

tensorflow tensorflow-serving tensorflow-estimator

share|improve this question

asked Mar 28 at 2:47

Xiang AplysiaXiang Aplysia

1255 bronze badges

add a comment | 

I want to export the model from checkpoint with estimator, and the variable data is stored in one file. I want the variable data to be stored in multi files so that I can distcp it faster.

I tried launched multi workers to export, but it does not work.

Here is some of my codes:

 if self.job_name == "worker":
 if FLAGS.ckpt != '' and int(FLAGS.ckpt) > 0:
 estimator.export_savedmodel(self.export_model_path,
 receiver_fn_raw(self.params['dense_num'], self.params['feature_conf'],
 self.params['list_num']),
 checkpoint_path=self.ckpt_dir)
 else:
 estimator.export_savedmodel(self.export_model_path,
 receiver_fn_raw(self.params['dense_num'], self.params['feature_conf'],
 self.params['list_num']))

tensorflow tensorflow-serving tensorflow-estimator

share|improve this question

asked Mar 28 at 2:47

Xiang AplysiaXiang Aplysia

1255 bronze badges

 if self.job_name == "worker":
 if FLAGS.ckpt != '' and int(FLAGS.ckpt) > 0:
 estimator.export_savedmodel(self.export_model_path,
 receiver_fn_raw(self.params['dense_num'], self.params['feature_conf'],
 self.params['list_num']),
 checkpoint_path=self.ckpt_dir)
 else:
 estimator.export_savedmodel(self.export_model_path,
 receiver_fn_raw(self.params['dense_num'], self.params['feature_conf'],
 self.params['list_num']))

share|improve this question

asked Mar 28 at 2:47

Xiang AplysiaXiang Aplysia

1255 bronze badges

share|improve this question

asked Mar 28 at 2:47

Xiang AplysiaXiang Aplysia

1255 bronze badges

asked Mar 28 at 2:47

Xiang AplysiaXiang Aplysia

1255 bronze badges

asked Mar 28 at 2:47

Xiang AplysiaXiang Aplysia

1255 bronze badges