Update app_dialogue.py

app_dialogue.py

@@ -2,11 +2,7 @@ import os
 import subprocess
 
 # Install flash attention
-subprocess.run(
-    "pip install flash-attn --no-build-isolation",
-    env={"FLASH_ATTENTION_SKIP_CUDA_BUILD": "TRUE"},
-    shell=True,
-)
+
 
 
 import copy
@@ -31,7 +27,7 @@ MODELS = {
     "idefics2-8b-chatty": Idefics2ForConditionalGeneration.from_pretrained(
         "HuggingFaceM4/idefics2-8b-chatty",
         torch_dtype=torch.bfloat16,
-        _attn_implementation="flash_attention_2",
+
     ).to(DEVICE),
 }
 PROCESSOR = AutoProcessor.from_pretrained(