Update app.py

app.py

@@ -1,6 +1,7 @@
 import gradio as gr
 import torch
 from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
+import time  # Untuk simulasi delay
 
 # Load the model and tokenizer
 model_name = "maulanayyy/codet5_code_translation-v3"  # Ganti dengan nama model yang benar
@@ -11,7 +12,7 @@ model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
 device = "cuda" if torch.cuda.is_available() else "cpu"
 model.to(device)  # Pindahkan model ke GPU jika tersedia
 
-# Function to perform inference
+# Function to perform inference with streaming
 def translate_code(input_code):
     try:
         # Prepare the input text
@@ -26,13 +27,18 @@ def translate_code(input_code):
         
         # Decode the output
         translated_code = tokenizer.decode(outputs[0], skip_special_tokens=True)
-        return translated_code
+        
+        # Simulate streaming output
+        for char in translated_code:
+            yield char  # Mengirimkan karakter satu per satu
+            time.sleep(0.05)  # Simulasi delay untuk streaming (dapat disesuaikan)
+        
     except Exception as e:
         print(f"Error during translation: {e}")
-        return "An error occurred during translation."
+        yield "An error occurred during translation."
 
-# Create Gradio interface
-demo = gr.Interface(fn=translate_code, inputs="text", outputs="text", title="Java to C# Code Translator", description="Enter Java code to translate it to C#.")
+# Create Gradio interface with streaming
+demo = gr.Interface(fn=translate_code, inputs="text", outputs="text", title="Java to C# Code Translator", description="Enter Java code to translate it to C#.", live=True)
 
 # Launch the interface
 demo.launch(share=True)