MultiHopKG
MultiHopKG copied to clipboard
salesforce
Multi-hop knowledge graph reasoning learned via policy gradient with reward shaping and action dropout
I was trying to run the model on my custom data of KG triples, to compare its performance, however I encountered a problem. Upon running the training command for policy...
error occurs when I train model using UMLS dataset Epoch 2: average training loss = -0.22322563640773296 entropy = 4.2052984714508055 => saving checkpoint to '/content/MultiHopKG-master/model/umls-point.rs.conve-xavier-n/a-200-200-3-0.001-0.1-0.0-0.95-400-0.05/checkpoint-2.tar' 0% 0/11 [00:00
The code only provides the .test file of the nell-995 data set. What is the difference between these two data sets? How to generate .test files for other data sets?...
pn.py L236 entity2bucketid = kg.entity2bucketid[e.tolist()] There is an error 【AttributeError: 'KnowledgeGraph' object has no attribute 'entity2bucketid'】 in the 236th line of pn.py, how can I solve it?
Hi there, Just as the title said, I don't know how to get the "raw.pgrk" file. Do you have the code? Thank you!
hi, browser your code, I don't find point model and point.gc model configs and lack the code about achieving the graph_convolution_layers. So, if you can update the code, I will...
Can you please explain whether it is possible to have relation and entities with different dimensions for doing reward shaping using Conve? I'm getting an assertion error when I have...
Hi, I tried action dropout trick on original minerva code on WN18RR. However, hit@10 decreased from 0.47 to 0.37 when action dropout rate changed from 1.0 to 0.9. Is there...
Metadata
Owner
Metadata
Multi-hop knowledge graph reasoning learned via policy gradient with reward shaping and action dropout