Avoid missing packages and attn_mask dtype error

FlagEmbedding/visual/modeling.py

@@ -1,16 +1,16 @@
-import os
 import logging
+import os
 from dataclasses import dataclass
 from typing import Optional, Tuple
+
 import torch
 import torch.distributed as dist
-from torch import nn, Tensor
-from transformers import AutoModel, AutoTokenizer, AutoConfig
+from PIL import Image
+from torch import Tensor, nn
+from transformers import AutoConfig, AutoModel, AutoTokenizer
 from transformers.file_utils import ModelOutput
 
-
 from FlagEmbedding.visual.eva_clip import create_eva_vision_and_transforms
-from PIL import Image
 
 logger = logging.getLogger(__name__)
 
@@ -200,6 +200,10 @@ def encode_text(self, texts):
             inputs_embeds=None,
             past_key_values_length=0,
         )
+
+        # Ensure the attention mask has the same dtype as the query tensor
+        extended_attention_mask = extended_attention_mask.to(embedding_output.dtype)
+
         encoder_outputs = self.bge_encoder(
             embedding_output,
             attention_mask=extended_attention_mask,

setup.py

@@ -1,22 +1,24 @@
-from setuptools import setup, find_packages
+from setuptools import find_packages, setup
 
 with open("README.md", mode="r", encoding="utf-8") as readme_file:
     readme = readme_file.read()
 
 setup(
-    name='FlagEmbedding',
-    version='1.2.10',
-    description='FlagEmbedding',
+    name="FlagEmbedding",
+    version="1.2.10",
+    description="FlagEmbedding",
     long_description=readme,
     long_description_content_type="text/markdown",
-    author_email='[email protected]',
-    url='https://github.com/FlagOpen/FlagEmbedding',
+    author_email="[email protected]",
+    url="https://github.com/FlagOpen/FlagEmbedding",
     packages=find_packages(),
     install_requires=[
-        'torch>=1.6.0',
-        'transformers>=4.33.0',
-        'datasets',
-        'accelerate>=0.20.1',
-        'sentence_transformers',
+        "torch>=1.6.0",
+        "transformers>=4.33.0",
+        "datasets",
+        "accelerate>=0.20.1",
+        "sentence_transformers",
+        "peft",
+        "sentencepiece",
     ],
 )