generated from mathias/template-go-web
chore(loop): 6 more iters — plateau at ~0.34-0.37 (1/6 kept)
Toy encoder near ceiling. 1 kept (val_vol_r2 0.3032→0.3442), 5 reverts. Consistent plateau = time to swap in TS-JEPA backbone (#3/#5). Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This commit is contained in:
@@ -47,10 +47,10 @@ class Encoder(nn.Module):
|
||||
super().__init__()
|
||||
self.net = nn.Sequential(
|
||||
nn.Flatten(),
|
||||
nn.Linear(win * 2, 128),
|
||||
nn.LayerNorm(128),
|
||||
nn.Linear(win * 2, 256),
|
||||
nn.LayerNorm(256),
|
||||
nn.GELU(),
|
||||
nn.Linear(128, emb)
|
||||
nn.Linear(256, emb)
|
||||
)
|
||||
|
||||
def forward(self, x):
|
||||
@@ -61,7 +61,7 @@ def main():
|
||||
(Xtr, ytr), (Xte, yte) = build()
|
||||
Xtr_t = torch.tensor(Xtr, device=dev)
|
||||
enc = Encoder(WINDOW, EMBED_DIM).to(dev)
|
||||
dec = nn.Sequential(nn.Linear(EMBED_DIM, 128), nn.GELU(), nn.Linear(128, WINDOW * 2)).to(dev)
|
||||
dec = nn.Sequential(nn.Linear(EMBED_DIM, 256), nn.GELU(), nn.Linear(256, WINDOW * 2)).to(dev)
|
||||
opt = torch.optim.Adam(list(enc.parameters()) + list(dec.parameters()), lr=LR)
|
||||
|
||||
for _ in range(EPOCHS): # SSL: masked reconstruction of the window
|
||||
|
||||
Reference in New Issue
Block a user