Danqing Wang

可以，这里提供的是官方ROUGE-1.5.5的调用

你好！我们强烈建议你在GPU服务器上运行我们的代码，并且最低有32g的内存。如果不行的话，你最好把数据集按照行分割撑小文件并且修改_ExampleSet_ (https://github.com/brxx122/HeterSumGraph/blob/master/module/dataloader.py) 来需要时再读取数据。 具体来说，把_readJson_() 从_init_() 移动到 _get_example_() 并且按照id来读取文件。你也可以自己实现类似batch的机制。此外别忘了把**cache**目录下已处理好的feature文件也相应划分好。 Hi! We strongly recommend you to run our code in a GPU server with a larger CPU memory (at least 32g). If...

根据错误信息，应该是把单个例子当作单独的文件太小了从而导致的错误。我想你应该把比如2000个例子合成一个chunk来当作一份文件。你可以参考别人的做法 https://github.com/nlpyang/PreSumm/blob/master/src/prepro/data_builder.py According to the error message, the error may be caused by making an example as one file, which is too small for the multiprocessing function. I think that...

Please refer to the **Data** Section in _README.md_ and modify the _PrepareDataset.sh_ by yourself.

Yes. The label refers to the index of the selected sentence in the text. Since the model focuses on extractive summarization, which extracts sentences from the document instead of generating...

There are some possible reasons for this: 1. Researchers may use different versions of ROUGE. We use the official ROUGE perl script which has been most widely used. However, the...

> > 可以，这里提供的是官方ROUGE-1.5.5的调用 > > 作者您好，请问训练的时候是使用的rouge1.0.0 (pip install rouge==1.0.0)吗？ROUGE-1.5.5只有在测试的时候才会用到对吗？ 是的

Sorry, due to the license, we cannot share the dataset directly. Please apply for it based on the link provided.

What ROUGE score did you get in CNN/DM? The issue you mentioned above has no problem with the single document summary (CNN/DM) dataset. If you cannot get the results reported...

The training details are attached in Appendix. Specifically, "All models are trained on a single GeForce RTX 2080 Ti GPU for about 5 epochs."