Removing standard model

app.py

@@ -17,12 +17,12 @@ freeze(quantized_model)
 # Define the inference function
 def generate_text(prompt):
     # Measure time and generate text for the normal model
-    start_time_normal = time.time()
-    inputs = tokenizer(prompt, return_tensors='pt')
-    outputs_normal = model.generate(**inputs, max_length=100, num_return_sequences=1)
-    generated_text_normal = tokenizer.decode(outputs_normal[0], skip_special_tokens=True)
-    end_time_normal = time.time()
-    response_time_normal = end_time_normal - start_time_normal
+    # start_time_normal = time.time()
+    # inputs = tokenizer(prompt, return_tensors='pt')
+    # outputs_normal = model.generate(**inputs, max_length=100, num_return_sequences=1)
+    # generated_text_normal = tokenizer.decode(outputs_normal[0], skip_special_tokens=True)
+    # end_time_normal = time.time()
+    # response_time_normal = end_time_normal - start_time_normal
 
     # Measure time and generate text for the quantized model
     start_time_quantized = time.time()
@@ -31,7 +31,7 @@ def generate_text(prompt):
     end_time_quantized = time.time()
     response_time_quantized = end_time_quantized - start_time_quantized
 
-    return (generated_text_normal, f"{response_time_normal:.2f} seconds",
+    return (#generated_text_normal, f"{response_time_normal:.2f} seconds",
             generated_text_quantized, f"{response_time_quantized:.2f} seconds")
 
 # Create a Gradio interface
@@ -39,8 +39,8 @@ iface = gr.Interface(
     fn=generate_text,
     inputs=gr.Textbox(lines=2, placeholder="Enter your prompt here..."),
     outputs=[
-        gr.Textbox(label="Generated Text (Normal Model)"),
-        gr.Textbox(label="Response Time (Normal Model)"),
+        # gr.Textbox(label="Generated Text (Normal Model)"),
+        # gr.Textbox(label="Response Time (Normal Model)"),
         gr.Textbox(label="Generated Text (Quantized Model)"),
         gr.Textbox(label="Response Time (Quantized Model)")
     ],