Update app.py

Remove unsupported optimizations

app.py

@@ -29,7 +29,6 @@ import gradio as gr
 
 import torch
 from transformers import TorchAoConfig, AutoModelForCausalLM, AutoTokenizer
-import torchao.quantization.pt2e as pt2e
 from torchao.quantization import Float8DynamicActivationFloat8WeightConfig, Float8WeightOnlyConfig
 
 # quant_config = Float8WeightOnlyConfig()
@@ -47,9 +46,6 @@ model = AutoModelForCausalLM.from_pretrained(
 
 gc.collect()
 
-pt2e.move_exported_model_to_eval(model)
-
-gc.collect()
 
 #########