guided_summarization
guided_summarization copied to clipboard
neulab
GSum: A General Framework for Guided Neural Abstractive Summarization
I have downloaded a data from: https://drive.google.com/file/d/1DN7ClZCCXsk2KegmC6t4ClBwtAf5galI/view and i am trying to train the bert model using this command: ! bash train.sh path_to_data model_path log_path -visible_gpus -1 and I have...
A big chunk of your paper talks about highlighted sentences vs keywords vs triples vs retrieved summaries. I'd appreciate it if you could provide the model checkpoints for the BertAbs...
Hi I am new to running the files on colab and tried running this file for my thesis. I have created all the train.bpe.source, train.bpe.target... files under the input_files_bpe folder....
请问bart的模型训练完成后，如何进行测试呢。我根据readme里面的例子传参，遇到很多参数问题，通过debug倒是改的差不多了，但是又出现request的connectionError。应该怎么修改呢？ 下面是我的z_test使用pycharm parameters传参： ../../data/test.source-target.source ../../data/test.source-target.z ../../result ./trained_model/cnndm checkpoint_best.pt ../../data 以下的报错信息： `Traceback (most recent call last): File "/root/anaconda3/envs/LJA_bart/lib/python3.6/site-packages/urllib3/connectionpool.py", line 710, in urlopen chunked=chunked, File "/root/anaconda3/envs/LJA_bart/lib/python3.6/site-packages/urllib3/connectionpool.py", line 386, in _make_request self._validate_conn(conn) File...
Hi, why was the first token missed in my inference output? I used your ways to preprocess all my own data and trained it based on your command. But when...
Hello, While testing the model and running z_tes.sh, I got this repeated error messages: guided_summarization/bart/fairseq/sequence_generator.py:1146: UserWarning: An output with one or more elements was resized since it had shape [7],...
Hello, thank you very much for your time to answer the questions. We have a script to extract oralce guidance signals regarding keywords. Do you recommend any pre-trained models available...
I met the **"ZeroDivisionError: float division by zero"** when I want to train the model with multi-gpu. And if only 1 gpu, the problem disappear but the training is too...
Hello, I see that you have provided a way for generating sentence guidance signals through the BERText Model, Could you give some pointers for generating other guidance keywords, for eg...
关于论文的小疑问
大佬你好！感谢你的分享，受益匪浅！ 我有点没看懂这个结果。请问这个Oracle结果是在测试集上使用Oracle生成signals吗？个人感觉在测试集上使用标签来生成guidance signal有一些牵强。。。 如果是测试集的Oracle的话，请问有没有在训练时分别使用Oracle和automatic predicted的结果对比呀。  不好意思打扰你
