Update app.py

app.py

@@ -1,18 +1,15 @@
 import os
 import torch
 import spaces
-import subprocess
 import gradio as gr
 from huggingface_hub import login
 from transformers import AutoProcessor, PaliGemmaForConditionalGeneration
 
-subprocess.run('pip install flash-attn --no-build-isolation', env={'FLASH_ATTENTION_SKIP_CUDA_BUILD': "TRUE"}, shell=True)
 login(os.environ.get("HF_TOKEN"))
 
 model_id = "google/paligemma-3b-mix-448"
 model = PaliGemmaForConditionalGeneration.from_pretrained(
     model_id, device_map={"": 0},
-    attn_implementation="flash_attention_2",
     torch_dtype=torch.bfloat16,
 )
 processor = AutoProcessor.from_pretrained(model_id)