Sync with upstream

.clabot

@@ -1,4 +1,5 @@
 {
-  "contributors": ["eyurtsev", "hwchase17", "nfcampos", "efriis", "jacoblee93", "dqbd", "kreneskyp", "adarsh-jha-dev", "harris", "baskaryan", "hinthornw", "bracesproul", "jakerachleff", "craigsdennis", "anhi"],
+  "contributors": ["eyurtsev", "hwchase17", "nfcampos", "efriis", "jacoblee93", "dqbd", "kreneskyp", "adarsh-jha-dev", "harris", "baskaryan", "hinthornw", "bracesproul", "jakerachleff", "craigsdennis", "anhi", "169", "LarchLiu", "PaulLockett", "RCMatthias", "jwynia", "majiayu000", "mpskex", "shivachittamuru", "sinashaloudegi", "sowsan"
+],
   "message": "Thank you for your pull request and welcome to our community. We require contributors to sign our Contributor License Agreement, and we don't seem to have the username {{usersWithoutCLA}} on file. In order for us to review and merge your code, please complete the Individual Contributor License Agreement here https://forms.gle/Ljhqvt9Gdi1N385W6 .\n\nThis process is done manually on our side, so after signing the form one of the maintainers will add you to the contributors list.\n\nFor more details about why we have a CLA and other contribution guidelines please see: https://github.com/langchain-ai/langserve/blob/main/CONTRIBUTING.md."
 }

.github/workflows/_pydantic_compatibility.yml

@@ -13,7 +13,7 @@ env:
 
 jobs:
   build:
-    timeout-minutes: 5
+    timeout-minutes: 10
     defaults:
       run:
         working-directory: ${{ inputs.working-directory }}

.github/workflows/langserve_ci.yml

@@ -47,7 +47,7 @@ jobs:
       working-directory: .
     secrets: inherit
   test:
-    timeout-minutes: 5
+    timeout-minutes: 10
     runs-on: ubuntu-latest
     defaults:
       run:

README.md

@@ -257,6 +257,18 @@ pip install "gigaserve[client]"
 
 GigaServe работает как с runnable-интерфейсами(написанным с помощью constructed via [LangChain Expression Language](https://python.langchain.com/docs/expression_language/)), так и с классическими цепочками (посредством наследования от `Chain`). Но следует учиывать, что некоторые входные схемы для устаревших цепочек могут быть некорректными или неполными и могут вызывать ошибки. Это можно предотвратить, если обновить аттрибут `input_schema` таких цепочек в LangChain.
 
+### Deploy to Azure 
+
+You can deploy to Azure using Azure Container Apps (Serverless):
+
+```
+az containerapp up --name [container-app-name] --source . --resource-group [resource-group-name] --environment  [environment-name] --ingress external --target-port 8001 --env-vars=OPENAI_API_KEY=your_key  
+```
+
+You can find more info [here](https://learn.microsoft.com/en-us/azure/container-apps/containerapp-up)
+
+### Deploy to GCP
+
 ## Добавление аутентификации
 
 О том как добавить аутентификацию на свой сервер GigaServe вы может узнать в разделах документации FastAPI, посвященных [безопасности](https://fastapi.tiangolo.com/tutorial/security/) и [использованию связующего ПО](https://fastapi.tiangolo.com/tutorial/middleware/).

examples/configurable_chain/server.py

@@ -63,7 +63,28 @@
 
 
 ###############################################################################
-#             EXAMPLE 2: Configure fields based on Request metadata           #
+#                EXAMPLE 2: Configure prompt based on RunnableConfig          #
+###############################################################################
+configurable_prompt = PromptTemplate.from_template(
+    "tell me a joke about {topic}."
+).configurable_alternatives(
+    ConfigurableField(
+        id="prompt",
+        name="Prompt",
+        description="The prompt to use. Must contain {topic}.",
+    ),
+    default_key="joke",
+    fact=PromptTemplate.from_template(
+        "tell me a fact about {topic} in {language} language."
+    ),
+)
+prompt_chain = configurable_prompt | model | StrOutputParser()
+
+add_routes(app, prompt_chain, path="/configurable_prompt")
+
+
+###############################################################################
+#             EXAMPLE 3: Configure fields based on Request metadata           #
 ###############################################################################
 
 

examples/configurable_retrieval/client.ipynb

@@ -0,0 +1,168 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Client\n",
+    "\n",
+    "Demo of a client interacting with a configurable retriever (see server code)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "You can interact with this via API directly"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{'output': [{'page_content': 'cats like fish',\n",
+       "   'metadata': {},\n",
+       "   'type': 'Document'},\n",
+       "  {'page_content': 'dogs like sticks', 'metadata': {}, 'type': 'Document'}],\n",
+       " 'callback_events': [],\n",
+       " 'metadata': {'run_id': 'f375cdf6-2848-4976-9565-f69e175c24ce'}}"
+      ]
+     },
+     "execution_count": 1,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "import requests\n",
+    "\n",
+    "inputs = {\"input\": \"cat\"}\n",
+    "response = requests.post(\"http://localhost:8000/invoke\", json=inputs)\n",
+    "\n",
+    "response.json()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "You can also interact with this via the RemoteRunnable interface (to use in other chains)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "from langserve import RemoteRunnable\n",
+    "\n",
+    "remote_runnable = RemoteRunnable(\"http://localhost:8000/\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Remote runnable has the same interface as local runnables"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "[Document(page_content='cats like fish'),\n",
+       " Document(page_content='dogs like sticks')]"
+      ]
+     },
+     "execution_count": 3,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "await remote_runnable.ainvoke(\"cat\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "[Document(page_content='cats like fish'),\n",
+       " Document(page_content='dogs like sticks')]"
+      ]
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "await remote_runnable.ainvoke(\"cat\", {\"configurable\": {\"collection_name\": \"Index 1\"}})"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "[Document(page_content='x_n+1=a * xn * (1-xn)')]"
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "await remote_runnable.ainvoke(\"cat\", {\"configurable\": {\"collection_name\": \"Index 2\"}})"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.9.6"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}

examples/configurable_retrieval/server.py

@@ -0,0 +1,126 @@
+#!/usr/bin/env python
+"""A more complex example that shows how to configure index name at run time."""
+from typing import Any, Iterable, List, Optional, Type
+
+from fastapi import FastAPI
+from langchain.embeddings import OpenAIEmbeddings
+from langchain.schema import Document
+from langchain.schema.embeddings import Embeddings
+from langchain.schema.retriever import BaseRetriever
+from langchain.schema.runnable import (
+    ConfigurableFieldSingleOption,
+    RunnableConfig,
+    RunnableSerializable,
+)
+from langchain.schema.vectorstore import VST
+from langchain.vectorstores import FAISS, VectorStore
+
+from langserve import add_routes
+from langserve.pydantic_v1 import BaseModel, Field
+
+vectorstore1 = FAISS.from_texts(
+    ["cats like fish", "dogs like sticks"], embedding=OpenAIEmbeddings()
+)
+
+vectorstore2 = FAISS.from_texts(["x_n+1=a * xn * (1-xn)"], embedding=OpenAIEmbeddings())
+
+
+app = FastAPI(
+    title="LangChain Server",
+    version="1.0",
+    description="Spin up a simple api server using Langchain's Runnable interfaces",
+)
+
+
+class UnderlyingVectorStore(VectorStore):
+    """This is a fake vectorstore for demo purposes."""
+
+    def __init__(self, collection_name: str) -> None:
+        """Fake vectorstore that has a collection name."""
+        self.collection_name = collection_name
+
+    def as_retriever(self) -> BaseRetriever:
+        if self.collection_name == "index1":
+            return vectorstore1.as_retriever()
+        elif self.collection_name == "index2":
+            return vectorstore2.as_retriever()
+        else:
+            raise NotImplementedError(
+                f"No retriever for collection {self.collection_name}"
+            )
+
+    def add_texts(
+        self,
+        texts: Iterable[str],
+        metadatas: Optional[List[dict]] = None,
+        **kwargs: Any,
+    ) -> List[str]:
+        raise NotImplementedError()
+
+    @classmethod
+    def from_texts(
+        cls: Type[VST],
+        texts: List[str],
+        embedding: Embeddings,
+        metadatas: Optional[List[dict]] = None,
+        **kwargs: Any,
+    ) -> VST:
+        raise NotImplementedError()
+
+    def similarity_search(
+        self, embedding: List[float], k: int = 4, **kwargs: Any
+    ) -> List[Document]:
+        raise NotImplementedError()
+
+
+class ConfigurableRetriever(RunnableSerializable[str, List[Document]]):
+    """Create a custom retriever that can be configured by the user.
+
+    This is an example of how to create a custom runnable that can be configured
+    to use a different collection name at run time.
+
+    Configuration involves instantiating a VectorStore with a collection name.
+    at run time, so the underlying vectorstore should be *cheap* to instantiate.
+
+    For example, it should not be making any network requests at instantiation time.
+
+    Make sure that the vectorstore you use meets this criteria.
+    """
+
+    collection_name: str
+
+    def invoke(
+        self, input: str, config: Optional[RunnableConfig] = None
+    ) -> List[Document]:
+        """Invoke the retriever."""
+        vectorstore = UnderlyingVectorStore(self.collection_name)
+        retriever = vectorstore.as_retriever()
+        return retriever.invoke(input, config=config)
+
+
+configurable_collection_name = ConfigurableRetriever(
+    collection_name="index1"
+).configurable_fields(
+    collection_name=ConfigurableFieldSingleOption(
+        id="collection_name",
+        name="Collection Name",
+        description="The name of the collection to use for the retriever.",
+        options={
+            "Index 1": "index1",
+            "Index 2": "index2",
+        },
+        default="Index 1",
+    )
+)
+
+
+class Request(BaseModel):
+    __root__: str = Field(default="cat", description="Search query")
+
+
+add_routes(app, configurable_collection_name.with_types(input_type=Request))
+
+if __name__ == "__main__":
+    import uvicorn
+
+    uvicorn.run(app, host="localhost", port=8000)

examples/conversational_retrieval_chain/server.py

@@ -87,12 +87,14 @@ class ChatHistory(BaseModel):
 
     chat_history: List[Tuple[str, str]] = Field(
         ...,
-        extra={"widget": {"type": "chat", "input": "question", "output": "answer"}},
+        extra={"widget": {"type": "chat", "input": "question"}},
     )
     question: str
 
 
-conversational_qa_chain = _inputs | _context | ANSWER_PROMPT | ChatOpenAI()
+conversational_qa_chain = (
+    _inputs | _context | ANSWER_PROMPT | ChatOpenAI() | StrOutputParser()
+)
 chain = conversational_qa_chain.with_types(input_type=ChatHistory)
 
 app = FastAPI(
@@ -104,7 +106,7 @@ class ChatHistory(BaseModel):
 # /invoke
 # /batch
 # /stream
-add_routes(app, chain)
+add_routes(app, chain, enable_feedback_endpoint=True)
 
 if __name__ == "__main__":
     import uvicorn