Update app.py

app.py

@@ -1,73 +1,94 @@
 import gradio as gr
 import numpy as np
 import pandas as pd
-import joblib
+import pickle
 
-# Load the trained models
-rf_fullstk = joblib.load('rf_hacathon_fullstk.pkl')
-rf_prodengg = joblib.load('rf_hacathon_prodengg.pkl')
-rf_mkt = joblib.load('rf_hacathon_mkt.pkl')
+# Load trained models
+with open('rf_hacathon_fullstk.pkl', 'rb') as f1:
+    rf_fullstk = pickle.load(f1)
+with open('rf_hacathon_prodengg.pkl', 'rb') as f2:
+    rf_prodengg = pickle.load(f2)
+with open('rf_hacathon_mkt.pkl', 'rb') as f3:
+    rf_mkt = pickle.load(f3)
 
-# Define prediction functions for each model
-def predict_fullstk(degree_p, internship, DSA, java):
-    new_data = pd.DataFrame({
-        'degree_p': degree_p,
-        'internship': internship,
-        'DSA': DSA,
-        'java': java,
-    }, index=[0])
-    prediction = rf_fullstk.predict(new_data)[0]
-    probability = rf_fullstk.predict_proba(new_data)[0][1]
-    return 'Placed' if prediction == 1 else 'Not Placed', probability
+# Define prediction function
+def predict_placed(degree_p, internship, DSA, java, management, leadership, communication, sales, model_name):
+    if model_name == 'Full Stack':
+        new_data = pd.DataFrame({
+            'degree_p': degree_p,
+            'internship': internship,
+            'DSA': DSA,
+            'java': java,
+            'management': 0,
+            'leadership': 0,
+            'communication': 0,
+            'sales': 0
+        }, index=[0])
+        model = rf_fullstk
+    elif model_name == 'Product Engineering':
+        new_data = pd.DataFrame({
+            'degree_p': degree_p,
+            'internship': internship,
+            'DSA': 0,
+            'java': 0,
+            'management': management,
+            'leadership': leadership,
+            'communication': 0,
+            'sales': 0
+        }, index=[0])
+        model = rf_prodengg
+    elif model_name == 'Marketing':
+        new_data = pd.DataFrame({
+            'degree_p': degree_p,
+            'internship': internship,
+            'DSA': 0,
+            'java': 0,
+            'management': 0,
+            'leadership': 0,
+            'communication': communication,
+            'sales': sales
+        }, index=[0])
+        model = rf_mkt
 
-def predict_prodengg(degree_p, internship, management, leadership):
-    new_data = pd.DataFrame({
-        'degree_p': degree_p,
-        'internship': internship,
-        'management': management,
-        'leadership': leadership,
-    }, index=[0])
-    prediction = rf_prodengg.predict(new_data)[0]
-    probability = rf_prodengg.predict_proba(new_data)[0][1]
-    return 'Placed' if prediction == 1 else 'Not Placed', probability
+    prediction = model.predict(new_data)
+    probability = model.predict_proba(new_data)[0][1]
 
-def predict_mkt(degree_p, internship, communication, sales):
-    new_data = pd.DataFrame({
-        'degree_p': degree_p,
-        'internship': internship,
-        'communication': communication,
-        'sales': sales,
-    }, index=[0])
-    prediction = rf_mkt.predict(new_data)[0]
-    probability = rf_mkt.predict_proba(new_data)[0][1]
-    return 'Placed' if prediction == 1 else 'Not Placed', probability
+    if prediction == 1:
+        result = 'Placed'
+        probability_message = f"You will be placed with a probability of {probability:.2f}"
+    else:
+        result = 'Not Placed'
+        probability_message = ""
 
-# Create input and output components for each model
-fullstk_inputs = [
-    gr.inputs.Number(label='Degree Percentage', min_value=0, max_value=100),
-    gr.inputs.Radio(label='Internship', choices=[0, 1]),
-    gr.inputs.Radio(label='DSA', choices=[0, 1]),
-    gr.inputs.Radio(label='Java', choices=[0, 1])
-]
-fullstk_output = gr.outputs.Label(num_top_classes=2, label='Placement Status')
+    return result, probability_message
 
-prodengg_inputs = [
+# Create Gradio interface
+inputs = [
     gr.inputs.Number(label='Degree Percentage', min_value=0, max_value=100),
     gr.inputs.Radio(label='Internship', choices=[0, 1]),
+    gr.inputs.Radio(label='Data Structures & Algorithms', choices=[0, 1]),
+    gr.inputs.Radio(label='Java', choices=[0, 1]),
     gr.inputs.Radio(label='Management Skills', choices=[0, 1]),
-    gr.inputs.Radio(label='Leadership Skills', choices=[0, 1])
+    gr.inputs.Radio(label='Leadership Skills', choices=[0, 1]),
+    gr.inputs.Radio(label='Communication Skills', choices=[0, 1]),
+    gr.inputs.Radio(label='Sales Skills', choices=[0, 1]),
+    gr.inputs.Dropdown(label='Model Name', choices=['Full Stack', 'Product Engineering', 'Marketing'])
 ]
-prodengg_output = gr.outputs.Label(num_top_classes=2, label='Placement Status')
 
-mkt_inputs = [
-    gr.inputs.Number(label='Degree Percentage', min_value=0, max_value=100),
-    gr.inputs.Radio(label='Internship', choices=[0, 1]),
-    gr.inputs.Radio(label='Communication Skills', choices=[0, 1]),
-    gr.inputs.Radio(label='Sales Skills', choices=[0, 1])
+outputs = [
+    gr.outputs.Textbox(label='Placement Result'),
+    gr.outputs.Textbox(label='Placement Probability')
 ]
-mkt_output = gr.outputs.Label(num_top_classes=2, label='Placement Status')
 
-# Create the Gradio app
-fullstk_interface = gr.Interface(
-    fn=predict_fullstk,
-    inputs=fullstk
+app = gr.Interface(
+    fn=predict_placed,
+    inputs=inputs,
+    outputs=outputs,
+    title='Placement Prediction',
+    description='Predict placement outcome based on given inputs',
+    allow_flagging=False
+)
+
+# Run the app
+if __name__ == '__main__':
+    app.run()