Update app.py

app.py

@@ -20,8 +20,8 @@ torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32
 models = {
     # 'microsoft/Florence-2-large-ft': AutoModelForCausalLM.from_pretrained('microsoft/Florence-2-large-ft', trust_remote_code=True).to(device).eval(),
     # 'microsoft/Florence-2-large': AutoModelForCausalLM.from_pretrained('microsoft/Florence-2-large', trust_remote_code=True).to(device).eval(),
-    'microsoft/Florence-2-base-ft': AutoModelForCausalLM.from_pretrained('microsoft/Florence-2-base-ft', trust_remote_code=True).to(device).eval(),
-    'microsoft/Florence-2-base': AutoModelForCausalLM.from_pretrained("microsoft/Florence-2-base", torch_dtype=torch_dtype, trust_remote_code=True).to(device) # AutoModelForCausalLM.from_pretrained('microsoft/Florence-2-base', trust_remote_code=True).to(device).eval(),
+    'microsoft/Florence-2-base-ft': AutoModelForCausalLM.from_pretrained('microsoft/Florence-2-base-ft', trust_remote_code=True, attn_implementation="eager").to(device).eval(),
+    'microsoft/Florence-2-base': AutoModelForCausalLM.from_pretrained("microsoft/Florence-2-base", torch_dtype=torch_dtype, trust_remote_code=True, attn_implementation="eager").to(device).eval()
 }
 
 processors = {