Update app.py

app.py

@@ -7,6 +7,8 @@ import matplotlib.pyplot as plt
 import matplotlib.patches as patches
 import random
 import numpy as np
+import subprocess
+subprocess.run('pip install flash-attn --no-build-isolation', env={'FLASH_ATTENTION_SKIP_CUDA_BUILD': "TRUE"}, shell=True)
 
 model_id = 'microsoft/Florence-2-large'
 model = AutoModelForCausalLM.from_pretrained(model_id, trust_remote_code=True).eval()