Update evo_model.py

evo_model.py

@@ -5,31 +5,22 @@ from torch.nn import TransformerEncoder, TransformerEncoderLayer
 class EvoTransformer(nn.Module):
     def __init__(self, vocab_size=30522, d_model=512, nhead=8, num_layers=6, dim_feedforward=1024, dropout=0.1):
         super(EvoTransformer, self).__init__()
+        self.embedding = nn.Embedding(vocab_size, d_model)
+        self.memory_token = nn.Parameter(torch.zeros(1, 1, d_model))
+        
+        encoder_layer = TransformerEncoderLayer(d_model=d_model, nhead=nhead, dim_feedforward=dim_feedforward, dropout=dropout)
+        self.transformer = TransformerEncoder(encoder_layer, num_layers=num_layers)
+        
+        self.memory_proj = nn.Linear(d_model, d_model)
+        # 🔥 self.norm removed
 
-        class Encoder(nn.Module):
-            def __init__(self):
-                super().__init__()
-                self.embedding = nn.Embedding(vocab_size, d_model)
-                self.memory_token = nn.Parameter(torch.zeros(1, 1, d_model))
-                
-                encoder_layer = TransformerEncoderLayer(d_model=d_model, nhead=nhead, dim_feedforward=dim_feedforward, dropout=dropout)
-                self.transformer = TransformerEncoder(encoder_layer, num_layers=num_layers)
-                
-                self.memory_proj = nn.Linear(d_model, d_model)
-                self.norm = nn.LayerNorm(d_model)
-
-            def forward(self, input_ids):
-                x = self.embedding(input_ids)
-                memory_token = self.memory_token.expand(x.size(0), -1, -1)
-                x = torch.cat([memory_token, x], dim=1)
-                x = self.transformer(x)
-                x = self.norm(x)
-                memory_output = self.memory_proj(x[:, 0])
-                return memory_output
+    def forward(self, input_ids):
+        x = self.embedding(input_ids)
+        memory_token = self.memory_token.expand(x.size(0), -1, -1)
+        x = torch.cat([memory_token, x], dim=1)
 
-        self.encoder = Encoder()
-        self.classifier = nn.Linear(d_model, 1)  # For binary choice
+        x = self.transformer(x)
+        # 🔥 x = self.norm(x) removed
 
-    def forward(self, input_ids):
-        x = self.encoder(input_ids)
-        return x
+        memory_output = self.memory_proj(x[:, 0])
+        return memory_output