fix sentence transformers export

Author: echarlaix

optimum/exporters/openvino/__main__.py

@@ -65,6 +65,7 @@ def main_export(
     compression_ratio: Optional[float] = None,
     stateful: bool = True,
     convert_tokenizer: bool = False,
+    library_name: Optional[str] = None,
     **kwargs_shapes,
 ):
     """
@@ -139,7 +140,9 @@ def main_export(
     original_task = task
     task = TasksManager.map_from_synonym(task)
     framework = TasksManager.determine_framework(model_name_or_path, subfolder=subfolder, framework=framework)
-    library_name = TasksManager.infer_library_from_model(model_name_or_path, subfolder=subfolder)
+    library_name = TasksManager.infer_library_from_model(
+        model_name_or_path, subfolder=subfolder, library_name=library_name
+    )
 
     if task == "auto":
         try:

optimum/exporters/openvino/convert.py

@@ -374,7 +374,6 @@ def export_pytorch(
         try:
             with config.patch_model_for_export(model, model_kwargs=model_kwargs):
                 check_dummy_inputs_are_allowed(model, dummy_inputs)
-
                 inputs = config.ordered_inputs(model)
                 input_names = list(inputs.keys())
                 output_names = list(config.outputs.keys())

optimum/intel/openvino/modeling.py

@@ -15,6 +15,7 @@
 import logging
 import os
 from pathlib import Path
+from tempfile import TemporaryDirectory
 from typing import Optional, Union
 
 import numpy as np
@@ -50,6 +51,7 @@
 
 from optimum.exporters import TasksManager
 
+from ...exporters.openvino import main_export
 from ..utils.import_utils import is_timm_available, is_timm_version
 from .modeling_base import OVBaseModel
 from .utils import _is_timm_ov_dir
@@ -411,6 +413,48 @@ def forward(
         )
         return BaseModelOutput(last_hidden_state=last_hidden_state)
 
+    @classmethod
+    def _from_transformers(
+        cls,
+        model_id: str,
+        config: PretrainedConfig,
+        use_auth_token: Optional[Union[bool, str]] = None,
+        revision: Optional[str] = None,
+        force_download: bool = False,
+        cache_dir: Optional[str] = None,
+        subfolder: str = "",
+        local_files_only: bool = False,
+        task: Optional[str] = None,
+        trust_remote_code: bool = False,
+        load_in_8bit: Optional[bool] = None,
+        load_in_4bit: Optional[bool] = None,
+        **kwargs,
+    ):
+        save_dir = TemporaryDirectory()
+        save_dir_path = Path(save_dir.name)
+
+        # If load_in_8bit is not specified then compression_option should be set to None and will be set by default in main_export depending on the model size
+        compression_option = "fp32" if load_in_8bit is not None else None
+
+        # OVModelForFeatureExtraction works with Transformers type of models, thus even sentence-transformers models are loaded as such.
+        main_export(
+            model_name_or_path=model_id,
+            output=save_dir_path,
+            task=task or cls.export_feature,
+            subfolder=subfolder,
+            revision=revision,
+            cache_dir=cache_dir,
+            use_auth_token=use_auth_token,
+            local_files_only=local_files_only,
+            force_download=force_download,
+            trust_remote_code=trust_remote_code,
+            compression_option=compression_option,
+            library_name="transformers",
+        )
+
+        config.save_pretrained(save_dir_path)
+        return cls._from_pretrained(model_id=save_dir_path, config=config, load_in_8bit=load_in_8bit, **kwargs)
+
 
 MASKED_LM_EXAMPLE = r"""
     Example of masked language modeling using `transformers.pipelines`: