Chao Shang

@maqy1995 Thanks for your help to response this question.

Hi Kiran, The ConvTransE uses the 1d convolution, as shown in the formula (6). This is a weighted sum operation. So it can keep the translational property. Thanks for your...

Hi, thanks for your question. You can output the entity and relation embeddings on both training and evaluation in models.py.

Hi, thank you so much for your questions. 1. We didn't use the normalization here. You can try to use it and see if it can improve the performance. 2....

Thanks for your message. Please check the answered issue. Others have reproduced the results. First, please run the latest version of code with more epochs (such as 1000 epochs). Second,...

Hi, I haven't checked here for a while. Sorry for the late reply. Based on my memory, we just calculate the node degree based on the graph we mentioned in...

Hi, thanks for your questions. The rel_rec and rel_send are the index (one hot) vectors that are used to extract the embeddings of sender and receiver: receivers = torch.matmul(self.rel_rec, x)...

> Hello! I'm sorry to bother you.I have run the code and tried to adjust the parameters base_lr,but I found that the experimental results were quite different from the paper.The...

Hi, thanks for your message. First, please note that we have updated the results in table 4, 5 and 6 with corresponding explanation in the appendix. Your results are very...

Hi, Thanks for your question. I haven't faced this problem before. Based on the log here, it seems that the size of one matrix is different from another matrix in...