alexisbellot
commented
2 years ago Hi, Thank you for your interest. This definitions of t adds the time stamps to the feature vector as it is potentially useful information. You could keep it or remove it, there is no theoretically grounded reason either way.
Alexis
De : Joshua Hahn @.> Envoyé : mardi 2 août 2022 17:26 À : alexisbellot/Synthetic-Controls-in-Continuous-Time @.> Cc : Subscribed @.***> Objet : [alexisbellot/Synthetic-Controls-in-Continuous-Time] Appending linspace to data (Issue #2)
Hello, I am a student at Columbia University currently working as a summer research assistant. I was exploring your NC-SC paper, and came across this repository to better understand how it was actually implemented. In data.py, I was wondering what the purpose of the variable t was. You define it as t = torch.linepsace(0,0,1,0,y.shape[1]), and you concatenate it to X (line 27 & 29). Thank you for your help!
— Reply to this email directly, view it on GitHubhttps://github.com/alexisbellot/Synthetic-Controls-in-Continuous-Time/issues/2, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AK2L2EUZMG34SZOMJ23D4CLVXFD4PANCNFSM55LY36IQ. You are receiving this because you are subscribed to this thread.Message ID: @.***>
Hello, I am a student at Columbia University currently working as a summer research assistant. I was exploring your NC-SC paper, and came across this repository to better understand how it was actually implemented. In data.py, I was wondering what the purpose of the variable t was. You define it as t = torch.linepsace(0,0,1,0,y.shape[1]), and you concatenate it to X (line 27 & 29). Thank you for your help!