Update app.py

app.py

@@ -1,5 +1,4 @@
 import streamlit as st
-import numpy as np
 import torch
 import torch.nn as nn
 import torch.nn.functional as F
@@ -15,6 +14,7 @@ def set_background(image_file):
         .stApp {{
             background-image: url(data:image/png;base64,{encoded_string});
             background-size: cover;
+            background-attachment: fixed;
         }}
         </style>
         """,
@@ -27,10 +27,10 @@ set_background('path_to_your_image.png')
 # Custom title with blue color
 st.markdown("<h1 style='color: blue;'>Placement Analysis</h1>", unsafe_allow_html=True)
 
-# Define the same model architecture
+# Define the ANN model architecture
 class ANN_Model(nn.Module):
     def __init__(self, input_cols=10, hidden1=20, hidden2=20, output=1):  
-        super().__init__()
+        super(ANN_Model, self).__init__()
         self.f_connected1 = nn.Linear(input_cols, hidden1)
         self.f_connected2 = nn.Linear(hidden1, hidden2)
         self.out = nn.Linear(hidden2, output)
@@ -42,10 +42,11 @@ class ANN_Model(nn.Module):
         return x
 
 # Load the model
-model = ANN_Model()  # Initialize the model
-model.load_state_dict(torch.load("ANN_model.pth"))  # Load saved weights
-model.eval()  # Set to evaluation mode
+model = ANN_Model()
+model.load_state_dict(torch.load("ANN_model.pth"))
+model.eval()
 
+# Create three columns for input fields
 col1, col2, col3 = st.columns(3)
 
 with col1:
@@ -57,7 +58,7 @@ with col1:
 
 with col2:
     AptitudeTestScore = st.number_input("Aptitude Test Score", min_value=0, max_value=100, step=1)
-    SoftSkillsRating = st.number_input("Soft Skills Rating", min_value=0, max_value=5)
+    SoftSkillsRating = st.number_input("Soft Skills Rating", min_value=0, max_value=5, step=1)
     ExtracurricularActivities = st.selectbox("Extracurricular Activities", ["Yes", "No"])
     ExtracurricularActivities = 1 if ExtracurricularActivities == "Yes" else 0
 
@@ -67,18 +68,20 @@ with col3:
     SSC_Marks = st.number_input("SSC Marks", min_value=0, max_value=100, step=1)
     HSC_Marks = st.number_input("HSC Marks", min_value=0, max_value=100, step=1)
 
-
 # Predict Button
 if st.button("Predict Placement"):
     # Prepare input for model
-    input_data = [cgpa, internships, projects, certifications, AptitudeTestScore, SoftSkillsRating,
-                  ExtracurricularActivities, PlacementTraining, SSC_Marks, HSC_Marks]
-    # Example: Making a prediction
+    input_data = torch.tensor([[
+        cgpa, internships, projects, certifications, AptitudeTestScore, SoftSkillsRating,
+        ExtracurricularActivities, PlacementTraining, SSC_Marks, HSC_Marks
+    ]], dtype=torch.float32)
+
+    # Make a prediction
     with torch.no_grad():
-        output = model(torch.tensor(input_data)).numpy()
+        output = model(input_data).item()
         if output >= 0.5:
             st.markdown(
-                f"""
+                """
                 <div style='background-color: #d4edda; padding: 10px; border-radius: 5px;'>
                     <h4 style='color: #155724;'>🎉 Congratulations! You are likely to get placed.</h4>
                 </div>
@@ -87,7 +90,10 @@ if st.button("Predict Placement"):
             )
         else:
             st.markdown(
-                f"""
-                <div style='background-color: #f8d7da;
-::contentReference[oaicite:2]{index=2} """
- 
+                """
+                <div style='background-color: #f8d7da; padding: 10px; border-radius: 5px;'>
+                    <h4 style='color: #721c24;'>⚠️ You might need to improve your profile for better chances.</h4>
+                </div>
+                """,
+                unsafe_allow_html=True
+            )