fix phi-3-vision import (#11129)

python/llm/src/ipex_llm/transformers/model.py

@@ -37,18 +37,21 @@
 # SOFTWARE.
 #
 
+import copy
+import torch
+import warnings
 import transformers
+from typing import List
+from unittest.mock import patch
 from transformers.configuration_utils import PretrainedConfig
-from .utils import extract_local_archive_file, \
+from transformers.dynamic_module_utils import get_imports
-    load_state_dict, \
+
-    get_local_shard_files, load_imatrix_data
 from ipex_llm.ggml.quantize import ggml_tensor_qtype, gguf_mixed_qtype
 from ipex_llm.utils.common import invalidInputError
 from ipex_llm.transformers.gguf.api import load_gguf_model
-import torch
+
-import warnings
+from .utils import logger, load_state_dict
-import copy
+from .utils import extract_local_archive_file, get_local_shard_files, load_imatrix_data
-from .utils import logger
 
 patched_training_mode = None
 
@@ -98,10 +101,19 @@ def _load_pre():
     GPTJModel.__init__ = gptj_model_new_init
 
 
+def patch_flash_attn_import(filename: str) -> List[str]:
+    """Work around for https://huggingface.co/microsoft/phi-1_5/discussions/72."""
+    imports = get_imports(filename)
+    if "flash_attn" in imports:
+        imports.remove("flash_attn")
+    return imports
+
+
 class _BaseAutoModelClass:
     HF_MODEL = None
 
     @classmethod
+    @patch("transformers.dynamic_module_utils.get_imports", patch_flash_attn_import)
     def from_pretrained(cls,
                         *args,
                         **kwargs):
@@ -492,6 +504,7 @@ class _BaseAutoModelClass:
         return model
 
     @classmethod
+    @patch("transformers.dynamic_module_utils.get_imports", patch_flash_attn_import)
     def load_low_bit(cls,
                      pretrained_model_name_or_path,
                      *model_args,