fix: update

Desc : update project

Project/Backend/Server_Node/controllers/marksCalculator.controller.js

@@ -5,6 +5,8 @@ export const marksCalculator = async (req, res) => {
     const targetClass = req.body.class;
     const { curriculumIndex, tutorialIndex } = req.params;
 
+    const status = "";
+
     // console.log(curriculumIndex, tutorialIndex);
 
     try {
@@ -15,13 +17,24 @@ export const marksCalculator = async (req, res) => {
                     console.error(error);
                     return res.status(500).json({ error: 'An error occurred' });
                 }
-                const [predictedClass, confidence] = stdout.trim().split(',');
+                const [predicted_class_name, confidence] = stdout.trim().split(',');
+
+                const parsedConfidence = parseFloat(confidence).toFixed(2);
+                
+                let status = "";
+
+                if (predicted_class_name === targetClass && parsedConfidence > 85) {
+                    status = "pass";
+                } else {
+                    status = "fail";
+                }
 
                 res.status(200).json({
                     code: "01",
                     result: {
-                        predictedClass,
-                        confidence: parseFloat(confidence).toFixed(2),
+                        predicted_class_name,
+                        confidence: parsedConfidence,
+                        status
                     }
                 });
             });

Project/Backend/Server_Node/prediction_config/C1T1/predict.py

@@ -56,6 +56,27 @@ class theCNN(nn.Module):
         return torch.softmax(self.output(x), axis=1)
 
 
+# Load the mapping dictionary
+class_mapping = {
+    0: 'Eight',
+    1: 'Eleven',
+    2: 'Fifty',
+    3: 'Five',
+    4: 'Four',
+    5: 'Fourteen',
+    6: 'Nine',
+    7: 'One',
+    8: 'Seven',
+    9: 'Six',
+    10: 'Ten',
+    11: 'Thirteen',
+    12: 'Thirty',
+    13: 'Three',
+    14: 'Twenty',
+    15: 'Two'
+}
+
+
 # Load the trained model
 model = theCNN()  # Instantiate your model class
 model.load_state_dict(torch.load(
@@ -85,8 +106,11 @@ with torch.no_grad():
     predicted_class = torch.argmax(outputs[0]).item()
     confidence = outputs[0][predicted_class].item()
 
+# Get the predicted class name using the mapping dictionary
+predicted_class_name = class_mapping[predicted_class]
+
 # Calculate similarity or confidence percentage
 # target_class = sys.argv[2].lower()  # Class name passed from the API
 similarity = 100 * confidence
 
-print(f"{predicted_class},{similarity:.2f}")
+print(f"{predicted_class_name},{similarity:.2f}")