Merge remote-tracking branch 'upstream/dev' into dev

src/backend/langflow/components/llms/AmazonBedrock.py

@@ -0,0 +1,45 @@
+from typing import Optional
+from langflow import CustomComponent
+from langchain.llms.bedrock import Bedrock
+from langchain.llms.base import BaseLLM
+
+
+class AmazonBedrockComponent(CustomComponent):
+    display_name: str = "Amazon Bedrock"
+    description: str = "LLM model from Amazon Bedrock."
+
+    def build_config(self):
+        return {
+            "model_id": {
+                "display_name": "Model Id",
+                "options": [
+                    "ai21.j2-grande-instruct",
+                    "ai21.j2-jumbo-instruct",
+                    "ai21.j2-mid",
+                    "ai21.j2-mid-v1",
+                    "ai21.j2-ultra",
+                    "ai21.j2-ultra-v1",
+                    "anthropic.claude-instant-v1",
+                    "anthropic.claude-v1",
+                    "anthropic.claude-v2",
+                    "cohere.command-text-v14",
+                ],
+            },
+            "credentials_profile_name": {"display_name": "Credentials Profile Name"},
+            "streaming": {"display_name": "Streaming", "field_type": "bool"},
+            "code": {"show": False},
+        }
+
+    def build(
+        self,
+        model_id: str = "anthropic.claude-instant-v1",
+        credentials_profile_name: Optional[str] = None,
+    ) -> BaseLLM:
+        try:
+            output = Bedrock(
+                credentials_profile_name=credentials_profile_name,
+                model_id=model_id,
+            )  # type: ignore
+        except Exception as e:
+            raise ValueError("Could not connect to AmazonBedrock API.") from e
+        return output

src/backend/langflow/components/retrievers/AmazonKendra.py

@@ -0,0 +1,48 @@
+from typing import Optional
+from langflow import CustomComponent
+from langchain.retrievers import AmazonKendraRetriever
+from langchain.schema import BaseRetriever
+
+
+class AmazonKendraRetrieverComponent(CustomComponent):
+    display_name: str = "Amazon Kendra Retriever"
+    description: str = "Retriever that uses the Amazon Kendra API."
+
+    def build_config(self):
+        return {
+            "index_id": {"display_name": "Index ID"},
+            "region_name": {"display_name": "Region Name"},
+            "credentials_profile_name": {"display_name": "Credentials Profile Name"},
+            "attribute_filter": {
+                "display_name": "Attribute Filter",
+                "field_type": "code",
+            },
+            "top_k": {"display_name": "Top K", "field_type": "int"},
+            "user_context": {
+                "display_name": "User Context",
+                "field_type": "code",
+            },
+            "code": {"show": False},
+        }
+
+    def build(
+        self,
+        index_id: str,
+        top_k: int = 3,
+        region_name: Optional[str] = None,
+        credentials_profile_name: Optional[str] = None,
+        attribute_filter: Optional[dict] = None,
+        user_context: Optional[dict] = None,
+    ) -> BaseRetriever:
+        try:
+            output = AmazonKendraRetriever(
+                index_id=index_id,
+                top_k=top_k,
+                region_name=region_name,
+                credentials_profile_name=credentials_profile_name,
+                attribute_filter=attribute_filter,
+                user_context=user_context,
+            )  # type: ignore
+        except Exception as e:
+            raise ValueError("Could not connect to AmazonKendra API.") from e
+        return output

src/backend/langflow/components/vectorstores/Chroma.py

@@ -14,7 +14,7 @@ class ChromaComponent(CustomComponent):
     A custom component for implementing a Vector Store using Chroma.
     """
 
-    display_name: str = "Chroma (Custom Component)"
+    display_name: str = "Chroma"
     description: str = "Implementation of Vector Store using Chroma"
     documentation = "https://python.langchain.com/docs/integrations/vectorstores/chroma"
     beta = True

src/backend/langflow/config.yaml

@@ -268,8 +268,8 @@ retrievers:
   # ZepRetriever:
   #   documentation: "https://python.langchain.com/docs/modules/data_connection/retrievers/integrations/zep_memorystore"
 vectorstores:
-  Chroma:
-    documentation: "https://python.langchain.com/docs/modules/data_connection/vectorstores/integrations/chroma"
+  # Chroma:
+  #   documentation: "https://python.langchain.com/docs/modules/data_connection/vectorstores/integrations/chroma"
   Qdrant:
     documentation: "https://python.langchain.com/docs/modules/data_connection/vectorstores/integrations/qdrant"
   Weaviate:

src/backend/langflow/interface/agents/custom.py

@@ -106,9 +106,9 @@ class CSVAgent(CustomAgentExecutor):
             tools,
             prefix=PANDAS_PREFIX,
             suffix=PANDAS_SUFFIX,
-            input_variables=["df", "input", "agent_scratchpad"],
+            input_variables=["df_head", "input", "agent_scratchpad"],
         )
-        partial_prompt = prompt.partial(df=str(df.head()))
+        partial_prompt = prompt.partial(df_head=str(df.head()))
         llm_chain = LLMChain(
             llm=llm,
             prompt=partial_prompt,

src/backend/langflow/interface/initialize/loading.py

@@ -300,6 +300,8 @@ def instantiate_embedding(node_type, class_object, params: Dict):
 
 def instantiate_vectorstore(class_object: Type[VectorStore], params: Dict):
     search_kwargs = params.pop("search_kwargs", {})
+    if search_kwargs == {"yourkey": "value"}:
+        search_kwargs = {}
     # clean up docs or texts to have only documents
     if "texts" in params:
         params["documents"] = params.pop("texts")