mniepert / mmkb Goto Github PK

Hi Alberto, I have been trying for a while to replicate the results from the time aware link prediction paper using the published datasets using the re-implementation of your models from this repo, https://github.com/INK-USC/RE-Net I have carefully looked into the code it looks reasonable to your implementation. The only change I have made from this code and my version is that I change the way they calculate filter hits@10 -> they suppress every object from the ranked list that is observed in training/validation regardless of the time when it occurred while I only suppress those observed in the same timestamp, otherwise I keep it.

The best results I have gotten so far are

  | MRR . | H1 | H10
TA-DistMult | 0.351600377 | 0.2187883521 | 0.628807319
TA-TransE | 0.2173162281 | 0 | 0.6098404552

Do you have any thoughts?? or maybe an accessible implementation of your model?

Thanks!

Dear Alberto and Mathias,

Is it possible for you to share the code of the paper [1]?
It seems that the current repository only stores the numerical KGs but not the code of embeddings?

Many thanks and best regards,
Daria

[1] Alberto García-Durán, Mathias Niepert: KBlrn: End-to-End Learning of Knowledge Base Representations with Latent, Relational, and Numerical Features. UAI 2018: 372-381

Hi,

In your paper, you said

Lastly, we have populated the three knowledge graphs with numeric literals and images for (almost) all of their entities.

So I'm wondering how you dealt with those missing images for some entities in your experiment for this paper?

i not only need the data,i need the code of your paper too,can you guys show me ?i need the TemporalKGS code ,that is to say code of this paper Learning Sequence Encoders for Temporal Knowledge Graph Completion

Could you please share the FB15k-num, FB15k-237-num used in the paper of "KBLRN: End-to-End Learning of Knowledge Base Representation with Latent, Relational, and Numerical Features"?

I tried to reproduce the experiment result of Table 4 in this paper, but I can not create the same set of valid and testing data as Table 1. (".. where numerical features are never used for the triples.")

Thank you very much and I am looking forward to hearing from you.

Best regards,

Phuc

Hi,
Thanks for the great work.
I was downloading the ICEWS05-15 dataset here, and I did some statistic analysis with this dataset.
I found that there are 10488 entities in my analysis, which is different from the number of entities shown in table 2 in your paper.
For the other statistics regarding ICEWS05-15, namely Relationships, #Facts, #Distinct TS and Time Span, the result of my analysis is as same as the values you have presented in table 2.
I am wondering if it is a typo in your paper?

Best Regards,
Zifeng

Hello. I think mmkb is an outstanding work and I want to follow this work. I found that the image urls of FB15K is released. Could you please release the image urls of DB15K and Yago15K?

hi, i found the Triples number of DB15K_EntityTriples.txt that do not repeat is 89197. And you say: "Relational Graph is 99,028 for DB15K" in paper "MMKG: Multi-Modal Knowledge Graphs".

For exmple,
<http://dbpedia.org/resource/Gwynedd> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://dbpedia.org/resource/Athlete> . appears four times in the file DB15K_EntityTriples.txt ,

Is there any negligence, or do i misidentity this?

I can't run the download-images.py to get the images, and it does not exist: b'image-graph_urls/URLS_google.txt'
Can you provide this url.txt file, thank your very much.