Update app.py

app.py

@@ -71,4 +71,30 @@ def ocr_and_extract(image, text_query):
         # Generate the output with Qwen2-VL
         generated_ids = qwen_model.generate(**inputs, max_new_tokens=50)
         output_text = processor.batch_decode(
-            generated_ids, skip_special_tokens=True, clean_up_tokeniza
+            generated_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False
+        )
+
+        # Clean up the temporary file
+        os.remove(temp_image_path)
+
+        return output_text[0]
+
+    except Exception as e:
+        error_message = str(e)
+        traceback.print_exc()
+        return f"Error: {error_message}"
+
+# Gradio interface for image input
+iface = gr.Interface(
+    fn=ocr_and_extract,
+    inputs=[
+        gr.Image(type="pil"),
+        gr.Textbox(label="Enter your query (optional)"),
+    ],
+    outputs="text",
+    title="Image OCR with Byaldi + Qwen2-VL",
+    description="Upload an image (JPEG/PNG) containing Hindi and English text for OCR.",
+)
+
+# Launch the Gradio app
+iface.launch()