Update app.py

app.py

@@ -1,33 +1,49 @@
-import gradio as gr
 import numpy as np
+import gradio as gr
 import joblib
 from pytorch_tabnet.tab_model import TabNetClassifier
 
-# Load model and preprocessing tools
+# Load model, scaler, and encoder
 model = TabNetClassifier()
-model.load_model("tabnet_model.zip")
-scaler = joblib.load("scaler.save")
-encoder = joblib.load("encoder.save")
+model.load_model('tabnet_model')
+
+scaler = joblib.load('D:/Dataset/IPIP-FFM-data-8Nov2018/scaler.save')
+encoder = joblib.load('D:/Dataset/IPIP-FFM-data-8Nov2018/encoder.save')
 
-# Features used in the model
-features = [f"{trait}{i}" for trait in ["EXT", "EST", "AGR", "CSN", "OPN"] for i in range(1, 11)]
+# Full form trait mapping
+trait_prefixes = {
+    'Extraversion': 'EXT',
+    'Emotional Stability': 'EST',
+    'Agreeableness': 'AGR',
+    'Conscientiousness': 'CSN',
+    'Openness': 'OPN'
+}
 
+# Create full feature names with full form labels
+feature_labels = []
+feature_keys = []
+for trait, abbrev in trait_prefixes.items():
+    for i in range(10):
+        feature_labels.append(f"{trait} Q{i+1}")
+        feature_keys.append(f"{abbrev}{i+1}")
+
+# Inference function
 def predict_personality(*inputs):
-    X = np.array(inputs).reshape(1, -1).astype(np.float32)
-    X_scaled = scaler.transform(X)
-    y_pred = model.predict(X_scaled)
-    label = encoder.inverse_transform(y_pred)[0]
-    return f"Predicted Personality Type: {label}"
+    input_array = np.array(inputs).reshape(1, -1)
+    scaled_input = scaler.transform(input_array)
+    pred = model.predict(scaled_input)
+    personality = encoder.inverse_transform(pred)[0]
+    return f"Predicted Personality: **{personality}**"
 
-# Create Gradio interface
-inputs = [gr.Slider(1, 5, step=0.1, label=f) for f in features]
+# Gradio UI: 50 sliders with full trait names
+inputs = [gr.Slider(1.0, 5.0, value=3.0, label=label) for label in feature_labels]
+output = gr.Textbox(label="Prediction")
 
 demo = gr.Interface(
     fn=predict_personality,
     inputs=inputs,
-    outputs=gr.Text(label="Personality Prediction"),
+    outputs=output,
     title="Personality Type Classifier (Introvert vs. Extrovert)",
-    description="This model predicts if a person is Introvert or Extrovert based on their IPIP-FFM scores."
+    description="Provide scores (1–5) for 50 questions from the IPIP-FFM questionnaire. The model will predict whether the person is an Introvert or an Extrovert."
 )
-
 demo.launch()