Styjun

How do I handle a DM that plays favorites with certain players?

how to change dot to underline in multiple file-names?

Why do my fried eggs start browning very fast?

What could prevent players from leaving an island?

How can I use commands with sudo without changing owner of the files?

The warlock of firetop mountain, what's the deal with reference 192?

A Checkmate of Dubious Legality

How easy is it to get a gun illegally in the United States?

If someone else uploads my GPL'd code to Github without my permission, is that a copyright violation?

Is there a booking app or site that lets you specify your gender for shared dormitories?

How do people drown while wearing a life jacket?

What does "autolyco-sentimental" mean?

How does Rust's 128-bit integer `i128` work on a 64-bit system?

Is an "are" omitted in this sentence

split inside flalign

What does C++ language definition say about the extent of the static keyword?

Can the Cauchy product of divergent series with itself be convergent?

Is it uncompelling to continue the story with lower stakes?

How to increase Solr JVM memory

How to call made-up data?

What license to choose for my PhD thesis?

Can a Hogwarts student refuse the Sorting Hat's decision?

What are the limitations of the Hendersson-Hasselbalch equation?

Awk to get all my regular users in shadow

Padding batch of uneven/unequal arrays in tensorflow

Nicest way to pad zeroes to a stringTensorflow: how to save/restore a model?Using a batchsize greater than 1 when using Tensorflows c++ APITensorFlow not found using pipTensorflow multiple samples without batchingTensorflow: LSTM and Convolution on subset of dimensions (avoiding unstack?)Tensorflow: building graph with batch sizes varying in dimension 1?Internally feeding to an intermediate tensorflow placeholderAttempted to use a closed SessionIndexing tensorflow output tensor in c++

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

0

I am trying to pad a batch of lists so that I get a rectangular tensor. The context is that I want to run batch training on unequal feature vectors. I believe there should be a simple for doing this in tensorflow but I am unable to find it.

I have already tried "pre-padding" the feature vectors before passing it using feed_dict and it works. However, I want to do this as a native Tensorflow operation as pre-padding in python might be slower than doing it in the graph.

def pad(train_features_list):
 max_seq_len = max(len(features) for features in train_features_list)
 for k, _ in enumerate(train_features_list):
 pads_to_add = max_seq_len - len(train_features_list[k])
 train_features_list[k].extend(['<PAD>'] * pads_to_add)

python tensorflow

share|improve this question

asked Mar 27 at 2:51

Rajhans SamdaniRajhans Samdani

111 bronze badge

add a comment | 

0

I am trying to pad a batch of lists so that I get a rectangular tensor. The context is that I want to run batch training on unequal feature vectors. I believe there should be a simple for doing this in tensorflow but I am unable to find it.

I have already tried "pre-padding" the feature vectors before passing it using feed_dict and it works. However, I want to do this as a native Tensorflow operation as pre-padding in python might be slower than doing it in the graph.

def pad(train_features_list):
 max_seq_len = max(len(features) for features in train_features_list)
 for k, _ in enumerate(train_features_list):
 pads_to_add = max_seq_len - len(train_features_list[k])
 train_features_list[k].extend(['<PAD>'] * pads_to_add)

python tensorflow

share|improve this question

asked Mar 27 at 2:51

Rajhans SamdaniRajhans Samdani

111 bronze badge

add a comment | 

I am trying to pad a batch of lists so that I get a rectangular tensor. The context is that I want to run batch training on unequal feature vectors. I believe there should be a simple for doing this in tensorflow but I am unable to find it.

I have already tried "pre-padding" the feature vectors before passing it using feed_dict and it works. However, I want to do this as a native Tensorflow operation as pre-padding in python might be slower than doing it in the graph.

def pad(train_features_list):
 max_seq_len = max(len(features) for features in train_features_list)
 for k, _ in enumerate(train_features_list):
 pads_to_add = max_seq_len - len(train_features_list[k])
 train_features_list[k].extend(['<PAD>'] * pads_to_add)

python tensorflow

share|improve this question

asked Mar 27 at 2:51

Rajhans SamdaniRajhans Samdani

111 bronze badge

def pad(train_features_list):
 max_seq_len = max(len(features) for features in train_features_list)
 for k, _ in enumerate(train_features_list):
 pads_to_add = max_seq_len - len(train_features_list[k])
 train_features_list[k].extend(['<PAD>'] * pads_to_add)

share|improve this question

asked Mar 27 at 2:51

Rajhans SamdaniRajhans Samdani

111 bronze badge

share|improve this question

asked Mar 27 at 2:51

Rajhans SamdaniRajhans Samdani

111 bronze badge

asked Mar 27 at 2:51

Rajhans SamdaniRajhans Samdani

111 bronze badge

asked Mar 27 at 2:51

Rajhans SamdaniRajhans Samdani

111 bronze badge