LFNA ok on the valid data

exps/LFNA/lfna.py

@@ -99,18 +99,13 @@ def online_evaluate(env, meta_model, base_model, criterion, args, logger):
         with torch.no_grad():
             meta_model.eval()
             base_model.eval()
-            _, [future_container], _ = meta_model(
+            _, [future_container], time_embeds = meta_model(
                 future_time.to(args.device).view(1, 1), None, True
             )
             future_x, future_y = future_x.to(args.device), future_y.to(args.device)
             future_y_hat = base_model.forward_with_container(future_x, future_container)
             future_loss = criterion(future_y_hat, future_y)
-            logger.log(
-                "[ONLINE] [{:03d}/{:03d}] loss={:.4f}".format(
-                    idx, len(env), future_loss.item()
-                )
-            )
-        refine = meta_model.adapt(
+        refine, post_refine_loss = meta_model.adapt(
             base_model,
             criterion,
             future_time.item(),
@@ -118,6 +113,13 @@ def online_evaluate(env, meta_model, base_model, criterion, args, logger):
             future_y,
             args.refine_lr,
             args.refine_epochs,
+            {"param": time_embeds, "loss": future_loss.item()},
+        )
+        logger.log(
+            "[ONLINE] [{:03d}/{:03d}] loss={:.4f}".format(
+                idx, len(env), future_loss.item()
+            )
+            + ", post-loss={:.4f}".format(post_refine_loss if refine else -1)
         )
     meta_model.clear_fixed()
     meta_model.clear_learnt()
@@ -244,21 +246,6 @@ def main(args):
     logger.log("The meta-model is\n{:}".format(meta_model))
 
     batch_sampler = EnvSampler(train_env, args.meta_batch, args.sampler_enlarge)
-    # train_env.reset_max_seq_length(args.seq_length)
-    # valid_env.reset_max_seq_length(args.seq_length)
-    valid_env_loader = torch.utils.data.DataLoader(
-        valid_env,
-        batch_size=args.meta_batch,
-        shuffle=True,
-        num_workers=args.workers,
-        pin_memory=True,
-    )
-    train_env_loader = torch.utils.data.DataLoader(
-        train_env,
-        batch_sampler=batch_sampler,
-        num_workers=args.workers,
-        pin_memory=True,
-    )
     pretrain_v2(base_model, meta_model, criterion, train_env, args, logger)
 
     # try to evaluate once
@@ -507,7 +494,7 @@ if __name__ == "__main__":
         help="The learning rate for the optimizer, during refine",
     )
     parser.add_argument(
-        "--refine_epochs", type=int, default=50, help="The final refine #epochs."
+        "--refine_epochs", type=int, default=40, help="The final refine #epochs."
     )
     parser.add_argument(
         "--early_stop_thresh",

exps/LFNA/lfna_meta_model.py

@@ -276,10 +276,10 @@ class LFNA_Meta(super_core.SuperModule):
     def forward_candidate(self, input):
         raise NotImplementedError
 
-    def adapt(self, base_model, criterion, timestamp, x, y, lr, epochs):
+    def adapt(self, base_model, criterion, timestamp, x, y, lr, epochs, init_info):
         distance = self.get_closest_meta_distance(timestamp)
         if distance + self._interval * 1e-2 <= self._interval:
-            return False
+            return False, None
         x, y = x.to(self._meta_timestamps.device), y.to(self._meta_timestamps.device)
         with torch.set_grad_enabled(True):
             new_param = self.create_meta_embed()
@@ -290,7 +290,11 @@ class LFNA_Meta(super_core.SuperModule):
             self.replace_append_learnt(timestamp, new_param)
             self.train()
             base_model.train()
-            best_new_param, best_loss = None, 1e9
+            if init_info is not None:
+                best_loss = init_info["loss"]
+                new_param.data.copy_(init_info["param"].data)
+            else:
+                best_new_param, best_loss = None, 1e9
             for iepoch in range(epochs):
                 optimizer.zero_grad()
                 _, [_], time_embed = self(timestamp.view(1, 1), None, True)
@@ -303,14 +307,14 @@ class LFNA_Meta(super_core.SuperModule):
                 loss.backward()
                 optimizer.step()
                 # print("{:03d}/{:03d} : loss : {:.4f} = {:.4f} + {:.4f}".format(iepoch, epochs, loss.item(), meta_loss.item(), match_loss.item()))
-                if loss.item() < best_loss:
+                if meta_loss.item() < best_loss:
                     with torch.no_grad():
-                        best_loss = loss.item()
+                        best_loss = meta_loss.item()
                         best_new_param = new_param.detach()
         with torch.no_grad():
             self.replace_append_learnt(None, None)
             self.append_fixed(timestamp, best_new_param)
-        return True
+        return True, meta_loss.item()
 
     def extra_repr(self) -> str:
         return "(_super_layer_embed): {:}, (_super_meta_embed): {:}, (_meta_timestamps): {:}".format(