PyTorch code for CoST: Contrastive Learning of Disentangled Seasonal-Trend Representations for Time Series Forecasting (ICLR 2022)
BSD 3-Clause "New" or "Revised" License
215
stars
43
forks
source link
Helloteacher, I really don’t understand this part: (rand_idx = np.random.randint(0, x_q.shape[1]); q_t = F.normalize(self.head_q(q_t[:, rand_idx]), dim=-1)). Why specifically extract a random time point instead of expanding over all time points? #26
I did not find that part in your paper.