Allowing HuggingFaceEmbeddings from the cached weight (#3084)

### https://github.com/hwchase17/langchain/issues/3079
Allow initializing HuggingFaceEmbeddings from the cached weight

langchain/embeddings/huggingface.py

@@ -1,5 +1,5 @@
 """Wrapper around HuggingFace embedding models."""
-from typing import Any, List
+from typing import Any, List, Optional
 
 from pydantic import BaseModel, Extra
 
@@ -29,6 +29,9 @@ class HuggingFaceEmbeddings(BaseModel, Embeddings):
     client: Any  #: :meta private:
     model_name: str = DEFAULT_MODEL_NAME
     """Model name to use."""
+    cache_folder: Optional[str] = None
+    """Path to store models. 
+    Can be also set by SENTENCE_TRANSFORMERS_HOME enviroment variable."""
 
     def __init__(self, **kwargs: Any):
         """Initialize the sentence_transformer."""
@@ -36,7 +39,9 @@ class HuggingFaceEmbeddings(BaseModel, Embeddings):
         try:
             import sentence_transformers
 
-            self.client = sentence_transformers.SentenceTransformer(self.model_name)
+            self.client = sentence_transformers.SentenceTransformer(
+                self.model_name, self.cache_folder
+            )
         except ImportError:
             raise ValueError(
                 "Could not import sentence_transformers python package. "