Update app.py

app.py

@@ -65,18 +65,21 @@ class MemoryEfficientNN(nn.Module):
         return self.layers(x.long())
 
 # Memory-efficient dataset
-class MemoryEfficientDataset(IterableDataset):
-    def __init__(self, X, y, batch_size):
-        self.X = X
-        self.y = torch.LongTensor(y)  # Convert labels to long tensors
-        self.batch_size = batch_size
-
-    def __iter__(self):
-        for i in range(0, len(self.y), self.batch_size):
-            X_batch = self.X[i:i+self.batch_size].toarray()
-            y_batch = self.y[i:i+self.batch_size]  # No need to add a new dimension
-            yield torch.FloatTensor(X_batch), y_batch
-# Train Memory-Efficient Neural Network
+class MemoryEfficientNN(nn.Module):
+    def __init__(self, input_size, hidden_size, num_classes):
+        super(MemoryEfficientNN, self).__init__()
+        self.layers = nn.Sequential(
+            nn.Embedding(input_size, hidden_size),
+            nn.ReLU(),
+            nn.Dropout(0.2),
+            nn.Linear(hidden_size, hidden_size),
+            nn.ReLU(),
+            nn.Dropout(0.2),
+            nn.Linear(hidden_size, num_classes)
+        )
+        
+    def forward(self, x):
+        return self.layers(x.long())
 X_train, X_test, y_train, y_test = train_test_split(contexts_encoded, emotions_target, test_size=0.2, random_state=42)
 input_size = X_train.shape[1]
 hidden_size = 64
@@ -269,6 +272,7 @@ def get_sentiment(text):
     result = sentiment_pipeline(text)[0]
     return f"Sentiment: {result['label']}, Score: {result['score']:.4f}"
 
+
 def process_input(text):
     try:
         normalized_text = normalize_context(text)
@@ -276,7 +280,8 @@ def process_input(text):
         
         rf_prediction = rf_model.predict(encoded_text)[0]
         isolation_score = isolation_forest.decision_function(encoded_text)[0]
-        nn_prediction = model(torch.LongTensor(encoded_text.toarray()).to(device)).argmax(dim=1).item()
+        nn_output = model(torch.LongTensor(encoded_text.toarray()).to(device))
+        nn_prediction = nn_output.argmax(dim=1).item()
         
         predicted_emotion = emotion_classes[rf_prediction]
         sentiment_score = isolation_score
@@ -299,7 +304,6 @@ def process_input(text):
         error_message = f"An error occurred: {str(e)}"
         print(error_message)  # Logging the error
         return error_message, error_message, error_message, error_message
-
 iface = gr.Interface(
     fn=process_input,
     inputs="text",