wassname/DeepTime
1
0
Fork 0
mirror of https://github.com/wassname/DeepTime.git synced 2026-07-14 11:13:31 +08:00
Code Issues Packages Projects Releases Wiki Activity

misc

Browse Source
This commit is contained in:
wassname
2022-11-22 15:07:42 +08:00
parent f62f6b7ff0
commit 00ae1d8b8f
4 changed files with 599 additions and 21 deletions
Download Patch File Download Diff File Expand all files Collapse all files
models/DeepTIMe.py
+7 -2
View File
@@ -36,18 +36,23 @@ class DeepTIMe(nn.Module):
def forward(self, x: Tensor, x_time: Tensor, y_time: Tensor) -> Tensor:
tgt_horizon_len = y_time.shape[1]
batch_size, lookback_len, _ = x.shape
# relative coordinates are the same for each batch, so we make them once and repeat them
coords = self.get_coords(lookback_len, tgt_horizon_len).to(x.device)
if y_time.shape[-1] != 0:
if y_time.shape[-1] != 0: # if y_time is not empty of features
time = torch.cat([x_time, y_time], dim=1)
coords = repeat(coords, '1 t 1 -> b t 1', b=time.shape[0])
coords = torch.cat([coords, time], dim=-1)
time_reprs = self.inr(coords)
else:
time_reprs = repeat(self.inr(coords), '1 t d -> b t d', b=batch_size)
a = self.inr(coords)
time_reprs = repeat(a, '1 t d -> b t d', b=batch_size)
print(coords.shape, a.shape, batch_size, time_reprs.shape)
lookback_reprs = time_reprs[:, :-tgt_horizon_len]
horizon_reprs = time_reprs[:, -tgt_horizon_len:]
w, b = self.adaptive_weights(lookback_reprs, x)
preds = self.forecast(horizon_reprs, w, b)
return preds