OpenNRE-PyTorch

OpenNRE-PyTorch copied to clipboard

qiubeixiansheng

CNN One

2


juanluis17

效果没有达到paper中报告的

1

您好，请问您跑出的模型达到了paper报告的效果吗，我跑出来好像低于paper中所报告的P/R curves.我用的数据也是paper中(train_instance=522611) 

ErxinYu

您好，请问能提供nyt10数据集的json格式下载连接吗？官网下的都是pb文件

hjc3613

PyTorch实现似乎比原paper的结果好？

4

我发现这个PyTorch的实现能取得比PCNN-ATT Paper里显著更好的 P@k 和 PR-Curves。不知道你跑的时候是不是也是这样？ 谢谢！

rivercold

RuntimeError: CUDNN_STATUS_EXECUTION_FAILED

1

Reading training data... Finish reading Reading testing data... Finish reading Initializing training model... Finish initializing Epoch 0 starts... Traceback (most recent call last): File "train.py", line 30, in con.train() File...

taufique74

您好！我写了一个BiLSTM encoder：  不知道图片有没有正常显示，代码如下： + class BiRNN(nn.Module): + def __init__(self, config): super(BiRNN, self).__init__() self.config = config self.sen_len = None self.in_width = self.config.word_size + 2 * self.config.pos_size self.rnn = nn.GRU(self.in_width, self.config.hidden_size,...

Anery

Question about how to build bags

1

Hi~ I noticed that you aggregate only **contiguous** instances with the same `tup` into the same bag, in other words, you aggregate instances **locally**. https://github.com/ShulinCao/OpenNRE-PyTorch/blob/master/gen_data.py#L152-L155 but RESIDE handles this by...

speedcell4

We have two attention matrix in `Class Selector` One is `relation_matrix`, the other one is `attention_matrix`. Both of them is a matrix with shape`(num_class, hidden_states)`. I have checked in OpenNRE...

liyucheng09

https://github.com/ShulinCao/OpenNRE-PyTorch/blob/master/networks/selector.py#L70 You use dropout after sen_matrix in the training stage, and not use it in the test stage. I think the nuance of dropout should be manipulated by model.eval(), rather...

WindChimeRan