Update app.py

app.py

@@ -1,43 +1,33 @@
 import gradio as gr
-import torch
-from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
+from transformers import pipeline
 
-model_name = "facebook/bart-large-cnn"
-
-tokenizer = AutoTokenizer.from_pretrained(model_name)
-model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
-
-device = "cuda" if torch.cuda.is_available() else "cpu"
-model = model.to(device)
+# Load summarization pipeline
+summarizer = pipeline(
+    "summarization",
+    model="facebook/bart-large-cnn"
+)
 
 def summarize_text(text):
     if not text.strip():
-        return "Please enter some text."
-
-    inputs = tokenizer(
+        return "Please enter some text to summarize."
+    
+    summary = summarizer(
         text,
-        return_tensors="pt",
-        truncation=True,
-        max_length=1024
-    ).to(device)
-
-    summary_ids = model.generate(
-        inputs["input_ids"],
         max_length=150,
-        min_length=50,
-        num_beams=4,
-        early_stopping=True
+        min_length=40,
+        do_sample=False
     )
-
-    return tokenizer.decode(summary_ids[0], skip_special_tokens=True)
-
-with gr.Blocks() as demo:
-    gr.Markdown("# 📄 BART Summarizer")
-
-    text_input = gr.Textbox(lines=10, label="Enter Text")
-    btn = gr.Button("Generate Summary")
-    output = gr.Textbox(label="Summary", lines=6)
-
-    btn.click(summarize_text, inputs=text_input, outputs=output)
-
-demo.launch()
+    
+    return summary[0]['summary_text']
+
+# Gradio Interface
+demo = gr.Interface(
+    fn=summarize_text,
+    inputs=gr.Textbox(lines=15, placeholder="Paste your text here..."),
+    outputs="text",
+    title="Advanced BART Text Summarizer",
+    description="Summarize long text using facebook/bart-large-cnn"
+)
+
+if __name__ == "__main__":
+    demo.launch()