Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / abisee/pointer-generator issues and pull requests

#100 - Question about the training loss.

Issue - State: open - Opened by caoxu915683474 over 6 years ago

#99 - about max_enc_steps

Issue - State: open - Opened by wangxuekui over 6 years ago - 2 comments

#98 - Professional Diagrams

Issue - State: closed - Opened by Ann1583 over 6 years ago

#97 - Continue training

Issue - State: open - Opened by jiangix01 over 6 years ago - 2 comments

#96 - OOV during decoding

Issue - State: open - Opened by alireza202 over 6 years ago - 5 comments

#95 - Multiple times article and result summary printed in debug mode

Issue - State: open - Opened by lakshyakh93 over 6 years ago - 1 comment

#94 - How to use word2vec or fasttext embeddings in this code?

Issue - State: closed - Opened by erammunawwar over 6 years ago - 1 comment

#93 - TF Flag, for mode is raising exception --mode="train"

Issue - State: open - Opened by Adityagrao over 6 years ago - 4 comments

#92 - Throws Exception during threading (last 4)

Issue - State: closed - Opened by Adityagrao over 6 years ago - 2 comments

#91 - Mode must be train/eval/decode

Issue - State: closed - Opened by Adityagrao over 6 years ago - 8 comments

#90 - How to summarize a text using pre-trained models?

Issue - State: open - Opened by yasersakkaf over 6 years ago - 8 comments

#89 - Anyone Change the Loss Function?

Issue - State: open - Opened by landmann over 6 years ago

#88 - Should we mask the attention before applying the softmax?

Issue - State: closed - Opened by eduOS over 6 years ago - 3 comments

#87 - pytorch implementation

Issue - State: open - Opened by pet67 over 6 years ago - 2 comments

#86 - Output node name of pretrained model

Issue - State: open - Opened by bangnguyen89 over 6 years ago

#85 - How to give input in the case of tensorflow serving?

Issue - State: open - Opened by Sharathnasa over 6 years ago - 2 comments

#84 - Question regarding dimension of weight w_c for coverage

Issue - State: closed - Opened by atulkum over 6 years ago

#83 - Some doubts about the function 'restore_best_model'

Issue - State: closed - Opened by Elvins over 6 years ago

#82 - memory problem

Issue - State: open - Opened by liujian19911023 over 6 years ago

#81 - Cloud I use Transfer learning for new data?

Issue - State: open - Opened by dondon2475848 over 6 years ago - 1 comment

#80 - attention mechanism using the new seq2seq library

Issue - State: open - Opened by JenningsL almost 7 years ago

#79 - Tensorflow 1.5

Issue - State: open - Opened by scigeek72 almost 7 years ago - 6 comments

#78 - bug in coverage loss?

Issue - State: closed - Opened by lipiji almost 7 years ago - 1 comment

#77 - How to use the pretrain model?

Issue - State: open - Opened by zhhengcs almost 7 years ago - 16 comments

#76 - Error while restarting the training for Coverage

Issue - State: closed - Opened by learningneo almost 7 years ago - 3 comments

#75 - Vocab file for pre-trained model

Issue - State: closed - Opened by ojncwyog almost 7 years ago - 1 comment

#74 - global_step/sec

Issue - State: open - Opened by zhangzhizz almost 7 years ago - 4 comments

#73 - difference between the pointer network used in nallapati 2016?

Issue - State: open - Opened by AlJohri almost 7 years ago

#72 - not sure about implementation of copy scatter function

Issue - State: open - Opened by jiacheng-xu almost 7 years ago - 5 comments

#71 - Using the trained model for text which does not have a ready made abstract

Issue - State: closed - Opened by learningneo almost 7 years ago - 2 comments

#70 - Is it more appropriate to evaluate the test results using SPICE?

Issue - State: open - Opened by eduOS almost 7 years ago

#69 - Question about GPU memory

Issue - State: open - Opened by jiangix01 almost 7 years ago

#68 - Error about 'Matrix already exists' under decode mode.

Issue - State: open - Opened by xiyan524 almost 7 years ago - 6 comments

#67 - question of attention_decoder() function

Issue - State: open - Opened by biandh almost 7 years ago

#66 - Best way to print out the attention weights

Issue - State: closed - Opened by talbaumel almost 7 years ago - 2 comments

#64 - Make compatible to TensorFlow past Nov 6, 2017

Pull Request - State: open - Opened by f0k almost 7 years ago - 2 comments

#62 - wrong for increasing hidden_dim,emb_dim

Issue - State: open - Opened by jiangix01 about 7 years ago - 1 comment

#61 - Output for the baseline model seems wrong after training for 962 steps

Issue - State: closed - Opened by LeenaShekhar about 7 years ago - 2 comments

#60 - Am I right in understanding the beam search in this process?

Issue - State: closed - Opened by eduOS about 7 years ago - 1 comment

#59 - length of enc_batch and dec_batch for axis=1

Issue - State: open - Opened by jiangix01 about 7 years ago - 2 comments

#58 - Multi-GPU support?

Issue - State: open - Opened by wjpjet about 7 years ago - 3 comments

#57 - Question about length of document and summary

Issue - State: open - Opened by jiangix01 about 7 years ago - 2 comments
Labels: question

#56 - Questions for p_gens

Issue - State: open - Opened by HanGuo97 about 7 years ago - 1 comment
Labels: question

#55 - Trained model availability?

Issue - State: closed - Opened by LeenaShekhar about 7 years ago - 1 comment

#54 - Can we minimize the negative log-likelihood of target words without pointer_gen mode?

Issue - State: closed - Opened by Sanqiang about 7 years ago - 2 comments
Labels: question

#53 - Question about generated summary

Issue - State: open - Opened by Opdoop about 7 years ago - 5 comments
Labels: question

#52 - Conversion code to python3

Pull Request - State: open - Opened by becxer about 7 years ago

#51 - Efficiently generate summaries from new data?

Issue - State: closed - Opened by ibarrien about 7 years ago - 9 comments

#49 - Tried running it on random internet news articles. Results look more extractive than abstractive?

Issue - State: open - Opened by anubhavmax about 7 years ago - 17 comments
Labels: duplicate, question

#48 - Getting different result when running the pretrained model

Issue - State: closed - Opened by bwang482 about 7 years ago - 1 comment

#47 - Coverage mechanism vs input feeding

Issue - State: open - Opened by TinaB19 about 7 years ago - 1 comment

#46 - is final_dist a probability distribution ?

Issue - State: closed - Opened by rajarsheem about 7 years ago - 1 comment
Labels: question

#45 - Question on the number of decoding sentences

Issue - State: open - Opened by haozheji about 7 years ago - 1 comment
Labels: question

#44 - The loss decreases too fast using LCSTS dataset

Issue - State: open - Opened by lixiang1991 about 7 years ago - 7 comments
Labels: question

#43 - After training how to summarize our own Text

Issue - State: closed - Opened by kergann about 7 years ago - 4 comments

#42 - The question about coverage loss

Issue - State: open - Opened by wjbianjason over 7 years ago - 3 comments
Labels: question

#41 - [Potential] error in computing attention and final word probability

Issue - State: closed - Opened by hashbangCoder over 7 years ago - 7 comments
Labels: bug

#40 - Can share the source code of Nallapati et al., 2016?

Issue - State: closed - Opened by lixiang1991 over 7 years ago - 1 comment
Labels: question

#39 - Clarification on paper and implementation

Issue - State: closed - Opened by hashbangCoder over 7 years ago - 2 comments
Labels: question

#38 - run on GPU

Pull Request - State: open - Opened by falcondai over 7 years ago

#37 - Can you share the Model file for testing on my side?

Issue - State: closed - Opened by JafferWilson over 7 years ago - 14 comments
Labels: question

#36 - It seems that text_format is undefined?

Issue - State: closed - Opened by eduOS over 7 years ago - 1 comment
Labels: question

#35 - Add info on where to get the rouge perl scripts

Pull Request - State: open - Opened by abhishekraok over 7 years ago

#34 - Fail to decode any meaningful output.

Issue - State: closed - Opened by ghost over 7 years ago - 5 comments
Labels: question

#33 - Errors while preparing the dataset

Issue - State: closed - Opened by LeenaShekhar over 7 years ago - 1 comment

#32 - Problem of the word embedding of OOV words of decoder inputs

Issue - State: closed - Opened by ghost over 7 years ago - 6 comments

#31 - Failed to load checkpoint on eval mode

Issue - State: closed - Opened by armancohan over 7 years ago - 3 comments

#30 - It seems that the checkpoint is problematic?

Issue - State: closed - Opened by eduOS over 7 years ago - 3 comments

#29 - When are the attention probabilities for repeated indices summed?

Issue - State: closed - Opened by Henry-E over 7 years ago - 2 comments
Labels: question

#28 - Final tensorflow loss for results reported in paper

Issue - State: open - Opened by Spandan-Madan over 7 years ago - 7 comments
Labels: question

#27 - calculation of p_final(w)

Issue - State: closed - Opened by lipiji over 7 years ago - 4 comments
Labels: question

#26 - scheduled-learning, temperature and diverse-beam-search and no NaN

Pull Request - State: open - Opened by udibr over 7 years ago

#25 - Out of vocabulary word vectors

Issue - State: closed - Opened by Henry-E over 7 years ago - 4 comments
Labels: question

#24 - Make Compatible with Python 3 & Windows.

Pull Request - State: open - Opened by abhishekraok over 7 years ago

#23 - Use best model for decoding

Pull Request - State: open - Opened by pltrdy over 7 years ago

#22 - Why not tf.scatter_nd_add but tf.scatter_nd in function _calc_final_dist?

Issue - State: closed - Opened by HsuWanTing over 7 years ago - 2 comments
Labels: question

#21 - Model producing largely extractive summaries

Issue - State: closed - Opened by hate5six over 7 years ago - 6 comments
Labels: question

#20 - Not overfitting by training and testing on the same dataset

Issue - State: closed - Opened by freesunshine0316 over 7 years ago
Labels: question

#19 - Training Step

Issue - State: open - Opened by ghost over 7 years ago - 2 comments
Labels: question

#18 - `run_summarization.py` isn't computing on GPU (but allocate memory)

Issue - State: open - Opened by pltrdy over 7 years ago - 15 comments
Labels: bug, question

#17 - FLAGS.pointer_gen in data.py

Issue - State: closed - Opened by makcbe over 7 years ago - 1 comment
Labels: bug

#16 - Failure to replicate results

Issue - State: closed - Opened by hate5six over 7 years ago - 28 comments
Labels: question

#15 - Create test

Pull Request - State: closed - Opened by Weili-NLP over 7 years ago

#14 - Note about chunking data

Issue - State: closed - Opened by abisee over 7 years ago

#13 - Building Graph: std::bad_alloc

Issue - State: closed - Opened by pltrdy over 7 years ago - 3 comments

#12 - Interpreting the results

Issue - State: open - Opened by raduk over 7 years ago - 23 comments
Labels: question

#11 - 'NoneType' object has no attribute 'enc_batch'

Issue - State: closed - Opened by sarmnoo over 7 years ago - 4 comments

#10 - License?

Issue - State: closed - Opened by abhishekraok over 7 years ago - 4 comments

#9 - Fixed `single_pass` related bugs of file and folder names.

Pull Request - State: closed - Opened by tianjianjiang over 7 years ago - 2 comments

#8 - add global tf.FLAGS variable to data.py

Pull Request - State: closed - Opened by blanche over 7 years ago - 1 comment

#7 - Why it outputs ". . . . . . . . . . . . ." when I run decode mode?

Issue - State: open - Opened by fishermanff over 7 years ago - 18 comments
Labels: question

#6 - Decoding Error

Issue - State: closed - Opened by scylla over 7 years ago - 3 comments
Labels: bug

#5 - Nan

Issue - State: closed - Opened by lizaigaoge550 over 7 years ago - 1 comment
Labels: duplicate

#4 - Get NAN loss after 35k steps

Issue - State: open - Opened by StevenLOL over 7 years ago - 31 comments
Labels: bug

#3 - Training speed

Issue - State: closed - Opened by bugtig over 7 years ago - 5 comments
Labels: question

#2 - Tensorboard vis error: vocab_metadata.tsv is not a file

Issue - State: closed - Opened by hate5six over 7 years ago - 2 comments

#1 - Training error: "TensorArray has size zero, but element shape <unknown> is not fully defined."

Issue - State: closed - Opened by RafaelMostert over 7 years ago - 11 comments
Labels: bug