-
Notifications
You must be signed in to change notification settings - Fork 49
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
some question about calculating probalitility #6
Comments
This paper didn't utilize the probability to predict the future event, it used the equation in the bottom of Page 7, please check. |
In practice, using the intensity function doesn't work very well. Utilising a non-linearity of the model output and then a regression layer works well for event type prediction. |
I also have the same question, and this paper states the conditional intensity function is used to predict the future event type and time stamp. However, I find the model in this repo only uses a linear layer to get the prediction result. I hope someone can help me, thanks a lot! |
it used the equation in the bottom of Page 7~ |
Thanks for your reply! I only find the equation in the bottom of Page 7 is used in Utils.log_likelihood function to get log-likelihood. And I am not sure if this equation is used in the prediction process. Maybe I miss or misunderstand some details. I will read the codes carefully. Thanks! |
Referring to the 3rd floor, the 5.3 experimental part of the paper mentions that Eq.7 has a general effect, and a linear layer is used for this |
Thanks a lot! |
I found that the intensity function is not used when calculating the probability,I don't know if you can answer for me. Thank you.
The text was updated successfully, but these errors were encountered: