diff --git a/src/models/train.py b/src/models/train.py
index c4a1731..68793b7 100644
--- a/src/models/train.py
+++ b/src/models/train.py
@@ -14,7 +14,7 @@ import torch.nn.functional as F
 from sklearn.metrics import accuracy_score, f1_score
 from torch.optim import AdamW
 from torch.optim.lr_scheduler import ReduceLROnPlateau
-from torch.utils.data import DataLoader, TensorDataset
+from torch.utils.data import DataLoader, TensorDataset, WeightedRandomSampler
 
 from src.models.lstm_attention import HeatRiskPredictor
 from src.utils.config import (
@@ -147,11 +147,21 @@ def train() -> HeatRiskPredictor:
     X_test_t = torch.tensor(X_test_np, dtype=torch.float32)
     y_test_t = torch.tensor(y_test_np, dtype=torch.long)
 
-    # -------------------- DataLoader --------------------
+    # -------------------- DataLoader (加权采样) --------------------
+    # 基于 y_short 的类别权重，解决极度不平衡问题
+    y_short_labels = y_train_np[:, 0]
+    class_counts = np.bincount(y_short_labels)
+    class_weights = 1.0 / class_counts
+    sample_weights = class_weights[y_short_labels]
+    sampler = WeightedRandomSampler(
+        weights=torch.from_numpy(sample_weights).float(),
+        num_samples=len(y_short_labels),
+        replacement=True,
+    )
     train_loader = DataLoader(
         TensorDataset(X_train_t, y_train_t),
         batch_size=BATCH_SIZE,
-        shuffle=True,
+        sampler=sampler,
     )
     val_loader = DataLoader(
         TensorDataset(X_val_t, y_val_t),