Update app.py

app.py

@@ -6,7 +6,19 @@ import numpy as np
 # Load your trained model
 model = joblib.load('trained_model.joblib')
 
-# Define grade to numeric conversion
+# Define all possible features the model expects
+ALL_INTERESTS = ['Research', 'Art', 'Cooking', 'Creativity', 'Technology', 
+                'Reading', 'Physics', 'Entrepreneurship', 'Public Speaking',
+                'Dancing', 'Mathematics', 'Playing Football', 'Problem-Solving',
+                'Writing', 'Music', 'Leadership']
+
+ALL_STRENGTHS = ['Logical Reasoning', 'Hands-on Skills', 'Detail-Oriented',
+                'Leadership', 'Innovative Thinking', 'Teamwork',
+                'Analytical Thinking', 'Communication', 'Creativity']
+
+OTHER_FEATURES = ['E-ICT', 'Economics', 'Government', 'Geography',
+                 'Business Management', 'Visual Arts', 'Literature']
+
 def grade_to_numeric(grade):
     if pd.isna(grade) or grade == "":
         return np.nan
@@ -16,10 +28,15 @@ def grade_to_numeric(grade):
     }
     return grade_map.get(grade, np.nan)
 
-def predict_career(desired_career, aggregate, english, core_maths, science, 
-                  social_studies, elective_maths, chemistry, physics, biology, interests):
-    # Create input dictionary
-    input_data = {
+def create_feature_vector(desired_career, aggregate, english, core_maths, science,
+                        social_studies, elective_maths, chemistry, physics, biology,
+                        interests, strengths):
+    # Create base dictionary with all features initialized to 0 or NaN
+    features = {f: 0 for f in ALL_INTERESTS + ALL_STRENGTHS}
+    features.update({f: np.nan for f in OTHER_FEATURES})
+    
+    # Add core academic features
+    features.update({
         "Desired_Career": desired_career,
         "Aggregate": aggregate,
         "English": english,
@@ -29,61 +46,71 @@ def predict_career(desired_career, aggregate, english, core_maths, science,
         "Elective Maths": elective_maths,
         "Chemistry": chemistry,
         "Physics": physics,
-        "Biology": biology,
-        "Interests": interests
-    }
+        "Biology": biology
+    })
+    
+    # Process interests
+    for interest in [i.strip() for i in interests.split(',')]:
+        interest_key = f"interest_{interest}"
+        if interest_key in features:
+            features[interest_key] = 1
+    
+    # Process strengths
+    for strength in [s.strip() for s in strengths.split(',')]:
+        strength_key = f"strength_{strength}"
+        if strength_key in features:
+            features[strength_key] = 1
     
     # Convert to DataFrame
-    student_df = pd.DataFrame([input_data])
+    df = pd.DataFrame([features])
     
     # Convert grades to numerical
     grade_cols = ['English', 'Core Maths', 'Science', 'Social Studies',
                  'Elective Maths', 'Chemistry', 'Physics', 'Biology']
     
     for col in grade_cols:
-        student_df[col] = student_df[col].apply(grade_to_numeric)
+        df[col] = df[col].apply(grade_to_numeric)
     
+    return df
+
+def predict_career(desired_career, aggregate, english, core_maths, science,
+                  social_studies, elective_maths, chemistry, physics, biology,
+                  interests, strengths):
     try:
+        # Create complete feature vector
+        input_df = create_feature_vector(
+            desired_career, aggregate, english, core_maths, science,
+            social_studies, elective_maths, chemistry, physics, biology,
+            interests, strengths
+        )
+        
         # Make prediction
-        probabilities = model.predict_proba(student_df)[0]
+        probabilities = model.predict_proba(input_df)[0]
         classes = model.classes_
         
-        # Get top 5 recommendations
+        # Get top recommendations
         top5_idx = np.argsort(probabilities)[::-1][:5]
         recommendations = [
-            (classes[i], float(probabilities[i])) 
+            (classes[i], float(probabilities[i]))
             for i in top5_idx
         ]
         
-        # Format output with analysis
+        # Format output
         output = "## Top Career Recommendations\n\n"
         for course, prob in recommendations:
             output += f"**{course}** ({prob*100:.1f}% match)\n"
         
-        # Add strengths analysis
-        output += "\n## Your Academic Strengths\n"
-        strong_subjects = []
-        if student_df['English'].iloc[0] <= 3:
-            strong_subjects.append("English")
-        if student_df['Core Maths'].iloc[0] <= 3:
-            strong_subjects.append("Mathematics")
-        if student_df['Science'].iloc[0] <= 3:
-            strong_subjects.append("Science")
-        
-        if strong_subjects:
-            output += f"- Excellent performance in: {', '.join(strong_subjects)}\n"
-        else:
-            output += "- Good overall academic performance\n"
-        
-        # Add interest alignment
-        output += f"\n## Interest Alignment\n- Your interests: {interests}\n"
+        output += "\n## Your Profile Highlights\n"
+        output += f"- Desired Career: {desired_career}\n"
+        output += f"- Key Interests: {interests}\n"
+        output += f"- Core Strengths: {strengths}\n"
         
         return output
         
     except Exception as e:
         return f"Error generating recommendations: {str(e)}"
 
-# Define Gradio interface
+# Gradio Interface
 with gr.Blocks(title="Career Path Predictor") as interface:
     gr.Markdown("# Career Recommendation System")
     gr.Markdown("Enter your academic details to get personalized career recommendations")
@@ -103,7 +130,11 @@ with gr.Blocks(title="Career Path Predictor") as interface:
             )
             interests = gr.Textbox(
                 label="Your Interests (comma separated)",
-                placeholder="e.g. programming, biology, creative writing"
+                placeholder="e.g. research, technology, leadership"
+            )
+            strengths = gr.Textbox(
+                label="Your Strengths (comma separated)",
+                placeholder="e.g. analytical thinking, teamwork"
             )
             
         with gr.Column():
@@ -122,13 +153,13 @@ with gr.Blocks(title="Career Path Predictor") as interface:
                 label="Science Grade",
                 value="B2"
             )
+            social_studies = gr.Dropdown(
+                ["A1", "B2", "B3", "C4", "C5", "C6", "D7", "E8", "F9"],
+                label="Social Studies Grade",
+                value="B2"
+            )
             
     with gr.Row():
-        social_studies = gr.Dropdown(
-            ["A1", "B2", "B3", "C4", "C5", "C6", "D7", "E8", "F9"],
-            label="Social Studies Grade",
-            value="B2"
-        )
         elective_maths = gr.Dropdown(
             ["A1", "B2", "B3", "C4", "C5", "C6", "D7", "E8", "F9"],
             label="Elective Maths Grade",
@@ -139,8 +170,6 @@ with gr.Blocks(title="Career Path Predictor") as interface:
             label="Chemistry Grade",
             value="B2"
         )
-        
-    with gr.Row():
         physics = gr.Dropdown(
             ["A1", "B2", "B3", "C4", "C5", "C6", "D7", "E8", "F9"],
             label="Physics Grade",
@@ -158,17 +187,21 @@ with gr.Blocks(title="Career Path Predictor") as interface:
     submit.click(
         fn=predict_career,
         inputs=[desired_career, aggregate, english, core_maths, science,
-                social_studies, elective_maths, chemistry, physics, biology, interests],
+                social_studies, elective_maths, chemistry, physics, biology,
+                interests, strengths],
         outputs=output
     )
     
     gr.Examples(
         examples=[
-            ["Medicine", 6, "A1", "A1", "A1", "A1", "A1", "A1", "B2", "A1", "biology, research"],
-            ["Computer Science", 9, "B2", "A1", "B2", "B3", "A1", "B2", "B2", "C4", "programming, math"]
+            ["Medicine", 6, "A1", "A1", "A1", "A1", "A1", "A1", "B2", "A1", 
+             "research, biology, leadership", "analytical thinking, detail-oriented"],
+            ["Computer Science", 9, "B2", "A1", "B2", "B3", "A1", "B2", "B2", "C4", 
+             "technology, problem-solving, mathematics", "logical reasoning, innovative thinking"]
         ],
         inputs=[desired_career, aggregate, english, core_maths, science,
-                social_studies, elective_maths, chemistry, physics, biology, interests]
+                social_studies, elective_maths, chemistry, physics, biology,
+                interests, strengths]
     )
 
 interface.launch()