Juncheng Liu

Really appreciate your reply! I got it. So the point is that ConvE can use only fewer parameters to achieve a similar performance compared with DistMult, right? I still have...

Hi, I used the data provided in the original repository. So, I am not sure whether the dataset statistic is the same as that mentioned in the paper. How you...

I got this error as well when I use PyTorch 1.10. After changing to PyTorch 1.8.1, everything is fine. You can take a look at [this issue](https://github.com/dbolya/yolact/issues/664). Seems related to...

And also how to reproduce the results in the paper?

> > Could you provide more detailed instructions for running the model? Thank you! > > `python training_model2_scai2.py --kg1f ./dbpedia/db_insnet_train.txt --kg2f ./dbpedia/db_onto_small_mini.txt --modelname ./models --method transe --bridge CMP-linear --alignf ./dbpedia/db_InsType_mini.txt`...

> @liu-jc How did you manage to evaluate the model? Evaluation script does not exist at all... I implemented the evaluation by myself. You can try to implement and see...

新加坡区的似乎也是没办法获取，不清楚是不是cgi的都不行了？

Hi @marcopeix, Could you please provide a sample .csv you used? We can look more into it.

Hi @gorold, Sorry, I don't quite understand. How these changes with torch.roll can help to fix the weights? We still have / dim in our codebase? https://github.com/SalesforceAIResearch/uni2ts/blob/27616d912b632fc77731b207e0b74970bed5870b/src/uni2ts/module/ts_embed.py#L178

Hi @dany4142 @Sample-design-alt, Could you provide more info about the issue? Are you trying to load the model trained by yourself?