Update app.py

app.py

@@ -2,9 +2,8 @@
 import os
 import math
 import requests
-from flask import Flask, request, jsonify
-from flask_cors import CORS
 from langdetect import detect
+import gradio as gr
 
 # -----------------------------------------------------------------------------
 # Configuration
@@ -12,14 +11,14 @@ from langdetect import detect
 HF_API_URL = "https://api-inference.huggingface.co/models/YOUR_USERNAME/YOUR_MODEL"
 HF_TOKEN = os.getenv("HF_TOKEN")
 
+if HF_TOKEN is None:
+    raise ValueError("HF_TOKEN environment variable not set!")
+
 HEADERS = {
     "Authorization": f"Bearer {HF_TOKEN}",
     "Content-Type": "application/json"
 }
 
-app = Flask(__name__)
-CORS(app)
-
 # -----------------------------------------------------------------------------
 # Utility Functions
 # -----------------------------------------------------------------------------
@@ -39,7 +38,6 @@ def normalize_labels(hf_output):
     result = {item["label"].lower(): float(item["score"]) for item in hf_output}
     human_p = result.get("human", 0.0)
     ai_p = result.get("ai", 0.0)
-
     return human_p, ai_p
 
 def hf_inference(text):
@@ -49,17 +47,14 @@ def hf_inference(text):
     return r.json()
 
 # -----------------------------------------------------------------------------
-# Core Endpoint
+# Gradio Prediction Function
 # -----------------------------------------------------------------------------
-@app.route("/analyze", methods=["POST"])
-def analyze():
-    data = request.get_json()
-    text = data.get("text", "").strip()
-
+def analyze_text(text):
+    text = text.strip()
     if not text:
-        return jsonify({"error": "Empty input"}), 400
+        return {"error": "Empty input"}
 
-    # 1. Language detection (supports linguistic auditing)
+    # 1. Language detection
     try:
         language = detect(text)
     except Exception:
@@ -67,9 +62,8 @@ def analyze():
 
     # 2. Hugging Face inference
     hf_raw = hf_inference(text)
-
     if not isinstance(hf_raw, list):
-        return jsonify({"error": "Unexpected model response", "raw": hf_raw}), 500
+        return {"error": "Unexpected model response", "raw": hf_raw}
 
     human_p, ai_p = normalize_labels(hf_raw)
 
@@ -80,7 +74,7 @@ def analyze():
     # 4. Epistemic uncertainty
     H = entropy([human_p, ai_p])
 
-    # 5. Explainability placeholder (XAI-ready schema)
+    # 5. Explainability placeholder
     explainability_stub = {
         "method": "pending",
         "note": (
@@ -91,7 +85,7 @@ def analyze():
         "token_attributions": []
     }
 
-    # 6. Fairness metadata (for downstream auditing)
+    # 6. Fairness metadata
     fairness_context = {
         "language": language,
         "human_probability": human_p,
@@ -122,25 +116,25 @@ def analyze():
         "fairness_audit_fields": fairness_context
     }
 
-    return jsonify(response)
+    return response
 
 # -----------------------------------------------------------------------------
-# Health Check
+# Gradio Interface
 # -----------------------------------------------------------------------------
-@app.route("/", methods=["GET"])
-def index():
-    return jsonify({
-        "system": "HATA API",
-        "capabilities": [
-            "Human vs AI classification",
-            "Probability calibration",
-            "Uncertainty estimation",
-            "Language-aware auditing",
-            "Explainability-ready schema",
-            "Fairness instrumentation"
-        ]
-    })
+iface = gr.Interface(
+    fn=analyze_text,
+    inputs=gr.Textbox(lines=5, placeholder="Enter text here..."),
+    outputs=gr.JSON(),
+    title="HATA: Human-AI Text Attribution",
+    description=(
+        "Detect whether text is human-written or AI-generated.\n"
+        "Supports uncertainty estimation, language-aware auditing, "
+        "and XAI-ready outputs."
+    )
+)
 
+# -----------------------------------------------------------------------------
+# Launch Gradio App
 # -----------------------------------------------------------------------------
 if __name__ == "__main__":
-    app.run(host="0.0.0.0", port=5000, debug=True)
+    iface.launch(server_name="0.0.0.0", server_port=7860)