Production Deploy: Minimal Gradio app with final configuration.

README.md

@@ -1,14 +1,7 @@
+
 ---
-title: Clickbait Prediction
-emoji: 👁
-colorFrom: gray
-colorTo: pink
+title: Clickbait Prediction Model
+emoji: 🚨
 sdk: gradio
-sdk_version: 5.49.1
-app_file: app.py
-pinned: false
-license: apache-2.0
-short_description: A lightweight and efficient NLP project
+python_version: 3.11 # CRITICAL FIX for TypeError: code expected 16, got 18
 ---
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,39 @@
+
+import gradio as gr
+from transformers import pipeline
+
+CLASSIFIER_MODEL_ID = "sks01dev/clickbait-classifier"
+
+# The pipeline loads assets directly from the Hub and handles pre/post-processing.
+classifier = pipeline(
+    "sentiment-analysis", 
+    model=CLASSIFIER_MODEL_ID,
+    tokenizer=CLASSIFIER_MODEL_ID,
+    return_all_scores=True 
+)
+
+def predict(headline):
+    # Runs inference and formats the output dictionary for Gradio Label
+    results = classifier(headline)[0]
+    
+    # Map generic LABEL_X to clear, human-readable output
+    # NOTE: results[0] is typically LABEL_0, results[1] is LABEL_1
+    formatted_output = {
+        "NOT CLICKBAIT (0)": results[0]['score'], 
+        "CLICKBAIT (1)": results[1]['score']      
+    }
+    return formatted_output
+
+# Gradio Interface Setup
+gr.Interface(
+    fn=predict, 
+    inputs=gr.Textbox(lines=2, label="Enter News Headline"),
+    outputs=gr.Label(num_top_classes=2, title="Prediction Confidence"),
+    title="World-Class Clickbait Predictor",
+    description="DeBERTa-v3-small model deployed for high-confidence headline analysis.",
+    examples=[
+        ["10 Ways To Instantly Improve Your Mood"],
+        ["Government Releases New Economic Policy Report"],
+        ["You Won't Believe What Happened When We Tested This!"],
+    ]
+).launch()

requirements.txt

@@ -0,0 +1,4 @@
+
+gradio
+transformers
+torch