Update app.py

app.py

@@ -27,8 +27,9 @@ import numpy as np
 from huggingface_hub import hf_hub_download, list_repo_files
 
 # flash attention
-# import subprocess
+import subprocess
 # subprocess.run('pip install flash-attn --no-build-isolation', env={'FLASH_ATTENTION_SKIP_CUDA_BUILD': "TRUE"}, shell=True)
+subprocess.run('pip install flash-attn --no-build-isolation', shell=True)
 
 # Define constants
 DESCRIPTION = "[ShowUI Demo](https://huggingface.co/showlab/ShowUI-2B)"
@@ -54,8 +55,8 @@ for file in files:
 model = Qwen2VLForConditionalGeneration.from_pretrained(
     destination_folder,
     torch_dtype=torch.bfloat16,
-    device_map="cpu",
-    # attn_implementation = "flash_attention_2",
+    # device_map="cpu",
+    attn_implementation = "flash_attention_2",
 )
 
 # Load the processor