Update app.py

app.py

@@ -81,7 +81,7 @@ usage_stats = {
 }
 
 @spaces.GPU
-def generate_text(prompt, max_length=256, temperature=0.7, top_p=0.9, top_k=150, num_beams=8, repetition_penalty=1.5, progress=gr.Progress()):
+def generate_text(prompt, max_length=256, temperature=0.7, top_p=0.9, top_k=150, num_beams=8, repetition_penalty=1.5):
     if not prompt.strip():
         logger.warning("Empty prompt submitted")
         return "", "الرجاء إدخال نص للتوليد (Please enter text to generate)"
@@ -91,29 +91,26 @@ def generate_text(prompt, max_length=256, temperature=0.7, top_p=0.9, top_k=150,
     
     start_time = time.time()
     
-    # Simply use the progress.tqdm wrapper for automatic progress tracking
-    with progress.tqdm("توليد النص (Generating text)", total=1) as pbar:
-        # Tokenize input
-        inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
-        
-        # Generate text
-        output = model.generate(
-            **inputs, 
-            max_length=max_length, 
-            temperature=temperature, 
-            top_p=top_p, 
-            do_sample=True,
-            repetition_penalty=repetition_penalty,
-            num_beams=num_beams,
-            top_k=top_k,
-            early_stopping=True,
-            pad_token_id=tokenizer.pad_token_id,
-            eos_token_id=tokenizer.eos_token_id,
-        )
-        
-        # Decode output
-        result = tokenizer.decode(output[0], skip_special_tokens=True)
-        pbar.update(1)
+    # Tokenize input
+    inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
+    
+    # Generate text
+    output = model.generate(
+        **inputs, 
+        max_length=max_length, 
+        temperature=temperature, 
+        top_p=top_p, 
+        do_sample=True,
+        repetition_penalty=repetition_penalty,
+        num_beams=num_beams,
+        top_k=top_k,
+        early_stopping=True,
+        pad_token_id=tokenizer.pad_token_id,
+        eos_token_id=tokenizer.eos_token_id,
+    )
+    
+    # Decode output
+    result = tokenizer.decode(output[0], skip_special_tokens=True)
     
     # Update stats
     generation_time = time.time() - start_time