search

rstriv / Know-Evolve

Implementation code for ICML '17 paper "Deep Temporal Reasoning for Dynamic Knowledge Graphs"
108 stars 22 forks source link

training detail and result #9

Open troy0215 opened 4 years ago

troy0215 commented 4 years ago

Hello, i implement you paper in Pytorch and there are some problems.

  1. How to process one entity is involved in tow succesive events when computing happened term loss? I compute in following way, is that correct? -log(t - t_bar+1e-5) +log(1e-5)
  2. I found the negtive log loss oscillated not decreased. And i saw the training result pictures in one issue before under this repository, the nll he displayed didn't decrease, too. Do you only care about the MAR and Hits@10 outperformance other baseline model, not consider the convergence of algorithm?

  1. I can't run your code, so do you remeber the training time when run your code? My code seems time comsuming... Thank you
CaiWZ commented 3 years ago

hello, if you implement paper in Pytorch, i don't know if you can share it with me ! thanks a lot!!