Stack Overflow
Stack Overflow

Questions tagged [textsum]

31 questions
1
vote
2 answers

Tensorflow - TextSum model: How to create my own training data

I am trying to create my own training data for TextSum model. As my understanding, I need to put my articles and abstracts to a binary file (in TFRecords). However, I can not create my own training data from raw text files. I don't understand format…
The Lazy Log
  • 3,564
  • 2
  • 20
  • 27
0
votes
0 answers

How can I refector example_pb2.Example() in tf 2.11.0

I write such code segments. import tensorflow as tf from tensorflow.core.example import example_pb2 # Write to tf.Example tf_example = example_pb2.Example() …
Lian Junhong
  • 1
  • 1
0
votes
0 answers

Why are multiple-gpu slower than single gpu

I trained textsum at 4-gpus and 1-gpu ,I found 4-gpus is slower than single gpu , and they all have low utilization, anyone can help me resolve this problem?
Jing Wang
  • 9
  • 1
0
votes
1 answer

How to prepare test data for textsum?

I have been able to successfully run the pre-trained model of TextSum (Tensorflow 1.2.1). The output consists of summaries of CNN & Dailymail articles (which are chuncked into bin format prior to testing). I have also been able to create the…
KRW4
  • 67
  • 2
  • 10
0
votes
0 answers

Textsum - full of stopwords in decoded results

I'm trying to implement the Tensorflow Textsum model on my own data. The training set contains around 50k articles. I didn't change any default settings in the textsum model. Currently, my 'average loss' is around 1.5 to 3.0. However, the decoded…
Di Zhu
  • 1
  • 1
0
votes
2 answers

Tensorflow serving error when connecting with client "input size does not match signature"

I have gone about exporting the textsum model using the export_textsum.py file shown below and when I connect using the textsumclient.py file below I receive the error: Traceback (most recent call last): File "textsum_client.py", line 90, in …
xtr33me
  • 936
  • 1
  • 13
  • 39
0
votes
1 answer

How to interpret this loss curve of textsum model?

I have been training the textsum seq2seq w/attention model for abstractive summarization on a training corpus of 600k articles + abstracts. Can this be regarded convergence? If so, can it be right that it converged after less than say 5k steps?…
anthnyprschka
  • 301
  • 3
  • 14
0
votes
0 answers

Tensorflow "Found input thread dead"

I have an error in the Tensorflow model "Textsum" (https://github.com/tensorflow/models/tree/master/textsum) When running bazel-bin/textsum/seq2seq_attention \ --mode=train \ --article_key=article \ --abstract_key=abstract \ …
Tom Higgins
  • 99
  • 2
  • 10
0
votes
1 answer

Tensorflow Serving - No versions of servable found under base path

I'm currently trying to use tensorflow serving to serve a trained "textsum" model. I am using TF 0.11, which after some reading, it seems automatically calls export_meta_graph which creates the exported files ckpt and ckpt.meta files. Under the…
xtr33me
  • 936
  • 1
  • 13
  • 39
0
votes
1 answer

Textsum model -Decode result is wrong

Does anyone test the textsum code to get the perfect result? Two days ago to see the message xtr33me made " #Textsum# - Incorrect decode results compared to ref file", I now test the situation and this is the same, I use the Chinese corpus,…
shengjin Lou
  • 11
  • 1
0
votes
0 answers

How to setup Textsum for TensorFlow serving

I am trying to setup the decode functionality of textsum using tensorflow serving but I haven't been able fully make sense of what is fully necessary to perform via the MNIST tutorial. Has anyone come across any other tutorials on setting up…
xtr33me
  • 936
  • 1
  • 13
  • 39
0
votes
1 answer

Approx how long should textsum training take to drop average loss to decent value

I am working on getting a textsum implementation working and recently got my own scraped data fed in. I started training last night against 38000 articles. This morning when I looked at the average loss, I was around 5.2000000. When I was playing…
xtr33me
  • 936
  • 1
  • 13
  • 39
0
votes
1 answer

How to format input data for textsum data_convert_example

I was hoping someone may be able to see where I am failing here. So I have scraped some data from buzzfeed and now I am trying to format a text file with which I can then send into data_convert_examples text_to_data formatter. I thought I had the…
xtr33me
  • 936
  • 1
  • 13
  • 39
0
votes
0 answers

textsum decode giving summaries of other articles as result

I training the model using the given toy data set. And to see the result real quick, I tried to test with just one article from toy dataset. When I tested with one articles, The decoder is giving the summary of other article as a result (other…
TUMU. S
  • 41
  • 3
-1
votes
1 answer

textsum does not converge

I have trained text sum for 5 days with the parameters recommended in the project page. I use a training set with more than 3 million article-summary pairs. At first running_average_loss decrease slowly from around 9 to around 4, but after that,…
xiuchen
  • 1
  • 2
1
2
3