Making File Compatible With Environments That Do Not Have Flash Attention

modeling_qwen.py

@@ -46,11 +46,13 @@ from transformers.utils import (
 )
 
 
-if is_flash_attn_2_available():
-    from flash_attn import flash_attn_func, flash_attn_varlen_func
-    from flash_attn.bert_padding import index_first_axis, pad_input, unpad_input  # noqa
+# if is_flash_attn_2_available():
+#     from flash_attn import flash_attn_func, flash_attn_varlen_func
+#     from flash_attn.bert_padding import index_first_axis, pad_input, unpad_input  # noqa
 
-    _flash_supports_window_size = "window_size" in list(inspect.signature(flash_attn_func).parameters)
+#     _flash_supports_window_size = "window_size" in list(inspect.signature(flash_attn_func).parameters)
+
+# REMOVE THE CODE ABOVE SHOULD ALLOW FOR RUNNING OF THIS MODEL WITHOUT FLASH ATTENTION SUPPORT.
 
 
 logger = logging.get_logger(__name__)