fixed errors

tasks/models/model_nn.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:45fab4b08e520f45893c39b8a94c1d444f59268393fd06635d3bb4d36b0b16e2
+size 439226

tasks/text.py

@@ -61,32 +61,50 @@ async def evaluate_text(request: TextEvaluationRequest):
     #--------------------------------------------------------------------------------------------
     # YOUR MODEL INFERENCE CODE HERE
 
-    #Load the embedding model
-    #model = SentenceTransformer("dunzhang/stella_en_400M_v5",trust_remote_code=True) 
+    # Set the device to MPS (if available)
+    device = torch.device("mps" if torch.backends.mps.is_available() else "cpu")
+    print(f"Using device: {device}")
+
     model_name = "sentence-transformers/paraphrase-multilingual-mpnet-base-v2"  # You can use other Sentence Transformers models as needed
     sentence_model = SentenceTransformer(model_name)
 
     # Convert each sentence into a vector representation (embedding)
-    embeddings = sentence_model.encode(test_dataset['quote'])
+    embeddings = sentence_model.encode(test_dataset['quote'], convert_to_tensor=True)
 
-    # Update the code below to replace the random baseline by your model inference within the inference pass where the energy consumption and emissions are tracked.
-    #--------------------------------------------------------------------------------------------   
-    
     # Make random predictions (placeholder for actual model inference)
     true_labels = test_dataset["label"]
 
-
-    #load the xgboost model
-    #with open("models/stella_400_xgb_500.pkl",'rb') as f:
+    # load the xgboost model
+    # with open("models/stella_400_xgb_500.pkl",'rb') as f:
     #    xgbclassifier = pickle.load(f)
-
-    model_nn = torch.load("models/model_nn.pth")
+    from torch import nn, optim
+
+    class SimpleNN2(nn.Module):
+        def __init__(self, input_dim, output_dim):
+            super(SimpleNN2, self).__init__()
+            self.fc1 = nn.Linear(input_dim, 128)  # Reduce hidden units
+            self.fc2 = nn.Linear(128, 64)  # Further reduce units
+            self.fc3 = nn.Linear(64, output_dim)
+            self.relu = nn.ReLU()
+            self.dropout = nn.Dropout(0.3)  # Add dropout
+            self.batch_norm1 = nn.BatchNorm1d(128)
+            self.batch_norm2 = nn.BatchNorm1d(64)
+
+        def forward(self, x):
+            x = self.relu(self.batch_norm1(self.fc1(x)))
+            x = self.dropout(x)  # Apply dropout
+            x = self.relu(self.batch_norm2(self.fc2(x)))
+            x = self.dropout(x)  # Apply dropout
+            x = self.fc3(x)  # Output raw logits
+            return x
+
+    model_nn = torch.load("../tasks/models/model_nn.pth", map_location=device)
 
     # Set the model to evaluation mode
     model_nn.eval()
 
-    #make inference
-    #predictions = xgbclassifier.predict(embeddings)
+    # make inference
+    # predictions = xgbclassifier.predict(embeddings)
 
     # Make predictions
     with torch.no_grad():