Add tests and separate embed and query instructions (#836)

@enoreyes @hwchase17 I have taken a stab at the cleanups. Note that in
my testing the separate query instruction was quite important (i.e.
performance suffered if you didn't separate them) - so I have put that
in to.

langchain/embeddings/huggingface.py

@@ -6,7 +6,11 @@
 from langchain.embeddings.base import Embeddings
 
 DEFAULT_MODEL_NAME = "sentence-transformers/all-mpnet-base-v2"
-DEFAULT_INSTRUCTION = "Represent the following text:"
+DEFAULT_INSTRUCT_MODEL = "hkunlp/instructor-large"
+DEFAULT_EMBED_INSTRUCTION = "Represent the document for retrieval: "
+DEFAULT_QUERY_INSTRUCTION = (
+    "Represent the question for retrieving supporting documents: "
+)
 
 
 class HuggingFaceEmbeddings(BaseModel, Embeddings):
@@ -80,15 +84,17 @@ class HuggingFaceInstructEmbeddings(BaseModel, Embeddings):
         .. code-block:: python
 
             from langchain.embeddings import HuggingFaceInstructEmbeddings
-            model_name = TODO
+            model_name = "hkunlp/instructor-large"
             hf = HuggingFaceInstructEmbeddings(model_name=model_name)
     """
 
     client: Any  #: :meta private:
-    model_name: str = DEFAULT_MODEL_NAME
+    model_name: str = DEFAULT_INSTRUCT_MODEL
     """Model name to use."""
-    instruction: str = DEFAULT_INSTRUCTION
-    """Instruction to use."""
+    embed_instruction: str = DEFAULT_EMBED_INSTRUCTION
+    """Instruction to use for embedding documents."""
+    query_instruction: str = DEFAULT_QUERY_INSTRUCTION
+    """Instruction to use for embedding query."""
 
     def __init__(self, **kwargs: Any):
         """Initialize the sentence_transformer."""
@@ -119,7 +125,7 @@ def embed_documents(self, texts: List[str]) -> List[List[float]]:
         """
         instruction_pairs = []
         for text in texts:
-            instruction_pairs.append([self.instruction, text])
+            instruction_pairs.append([self.embed_instruction, text])
         embeddings = self.client.encode(instruction_pairs)
         return embeddings.tolist()
 
@@ -132,6 +138,6 @@ def embed_query(self, text: str) -> List[float]:
         Returns:
             Embeddings for the text.
         """
-        instruction_pair = [self.instruction, text]
+        instruction_pair = [self.query_instruction, text]
         embedding = self.client.encode([instruction_pair])[0]
         return embedding.tolist()

tests/integration_tests/embeddings/test_huggingface.py

@@ -1,7 +1,10 @@
 """Test huggingface embeddings."""
 import unittest
 
-from langchain.embeddings.huggingface import HuggingFaceEmbeddings
+from langchain.embeddings.huggingface import (
+    HuggingFaceEmbeddings,
+    HuggingFaceInstructEmbeddings,
+)
 
 
 @unittest.skip("This test causes a segfault.")
@@ -21,3 +24,20 @@ def test_huggingface_embedding_query() -> None:
     embedding = HuggingFaceEmbeddings()
     output = embedding.embed_query(document)
     assert len(output) == 768
+
+
+def test_huggingface_instructor_embedding_documents() -> None:
+    """Test huggingface embeddings."""
+    documents = ["foo bar"]
+    embedding = HuggingFaceInstructEmbeddings()
+    output = embedding.embed_documents(documents)
+    assert len(output) == 1
+    assert len(output[0]) == 768
+
+
+def test_huggingface_instructor_embedding_query() -> None:
+    """Test huggingface embeddings."""
+    query = "foo bar"
+    embedding = HuggingFaceInstructEmbeddings()
+    output = embedding.embed_query(query)
+    assert len(output) == 768