From 85087aecd83ebb020b14906640b7b65e2a58a4d4 Mon Sep 17 00:00:00 2001 From: Richard Anthony Hein Date: Fri, 6 Sep 2024 06:54:14 +0000 Subject: [PATCH] update model --- playground/agents/use_cases/weather/main.py | 2 +- playground/demos/chatbot/chatbot-ui | 2 +- playground/demos/chatbot/server/__init__.py | 4 ---- .../demos/chatbot/server/dockerRunVllm.sh | 2 +- playground/demos/chatbot/server/server.py | 21 ++++++++++--------- .../demos/chatbot/server/server_models.py | 4 ++-- playground/demos/vLLM/vLLM_example.py | 2 +- 7 files changed, 17 insertions(+), 20 deletions(-) diff --git a/playground/agents/use_cases/weather/main.py b/playground/agents/use_cases/weather/main.py index 4ccdccce..0683aa5b 100644 --- a/playground/agents/use_cases/weather/main.py +++ b/playground/agents/use_cases/weather/main.py @@ -12,7 +12,7 @@ weather_api_key= "af6ef989b5c50a91ca068cc00df125b7", # Replace with your weathe llm = OpenAIChatLLM( base_url=api_base, api_key=api_key, - model="NousResearch/Meta-Llama-3-8B-Instruct", + model="NousResearch/Meta-Llama-3.1-8B-Instruct", temperature=0, streaming=False ) diff --git a/playground/demos/chatbot/chatbot-ui b/playground/demos/chatbot/chatbot-ui index 9d23e948..bc222331 160000 --- a/playground/demos/chatbot/chatbot-ui +++ b/playground/demos/chatbot/chatbot-ui @@ -1 +1 @@ -Subproject commit 9d23e948f0a64c2e8aa9cdb6a5020222f7b4d325 +Subproject commit bc2223311c531b40b3b8e42cac743eda325947e6 diff --git a/playground/demos/chatbot/server/__init__.py b/playground/demos/chatbot/server/__init__.py index 3d7154bd..8b137891 100644 --- a/playground/demos/chatbot/server/__init__.py +++ b/playground/demos/chatbot/server/__init__.py @@ -1,5 +1 @@ -from playground.demos.chatbot.server.vector_store import VectorStorage -__all__ = [ - "VectorStorage", -] \ No newline at end of file diff --git a/playground/demos/chatbot/server/dockerRunVllm.sh b/playground/demos/chatbot/server/dockerRunVllm.sh index ace21b2a..7e346b4d 100644 --- a/playground/demos/chatbot/server/dockerRunVllm.sh +++ b/playground/demos/chatbot/server/dockerRunVllm.sh @@ -5,4 +5,4 @@ docker run --runtime nvidia --gpus all \ --network=host \ --name vllm \ vllm/vllm-openai:latest \ - --model NousResearch/Meta-Llama-3-8B-Instruct \ No newline at end of file + --model NousResearch/Meta-Llama-3.1-8B-Instruct \ No newline at end of file diff --git a/playground/demos/chatbot/server/server.py b/playground/demos/chatbot/server/server.py index 62ca1655..a94def0b 100644 --- a/playground/demos/chatbot/server/server.py +++ b/playground/demos/chatbot/server/server.py @@ -2,20 +2,21 @@ import logging import os from urllib.parse import urlparse -from swarms.structs.agent import Agent + from dotenv import load_dotenv from fastapi import FastAPI, HTTPException, Request from fastapi.middleware.cors import CORSMiddleware from fastapi.responses import JSONResponse from fastapi.routing import APIRouter from fastapi.staticfiles import StaticFiles -from huggingface_hub import login -from swarms.prompts.chat_prompt import Message, Role -from swarms.prompts.conversational_RAG import QA_PROMPT_TEMPLATE_STR from playground.demos.chatbot.server.responses import StreamingResponse from playground.demos.chatbot.server.server_models import ChatRequest from playground.demos.chatbot.server.vector_storage import RedisVectorStorage from swarms.models.popular_llms import OpenAIChatLLM +# from huggingface_hub import login +from swarms.prompts.chat_prompt import Message, Role +from swarms.prompts.conversational_RAG import QA_PROMPT_TEMPLATE_STR +from swarms.structs.agent import Agent logging.basicConfig(level=logging.ERROR) @@ -26,9 +27,9 @@ dotenv_path = os.path.join( ) load_dotenv(dotenv_path) -hf_token = os.environ.get( - "HUGGINFACEHUB_API_KEY" -) # Get the Huggingface API Token +# hf_token = os.environ.get( +# "HUGGINFACEHUB_API_KEY" +# ) # Get the Huggingface API Token uploads = os.environ.get( "UPLOADS" @@ -47,7 +48,7 @@ openai_api_base = ( ) env_vars = [ - hf_token, + # hf_token, uploads, openai_api_key, openai_api_base, @@ -70,7 +71,7 @@ print(f"USE_GPU={use_gpu}") print(f"OPENAI_API_KEY={openai_api_key}") print(f"OPENAI_API_BASE={openai_api_base}") print("Logging in to huggingface.co...") -login(token=hf_token) # login to huggingface.co +# login(token=hf_token) # login to huggingface.co app = FastAPI(title="Chatbot") @@ -176,7 +177,7 @@ async def create_chat( # add docs to short term memory for data in [doc["content"] for doc in docs]: - agent.add_message_to_memory(role=Role.HUMAN, content=data) + agent.add_message_to_memory(data) async for response in agent.run_async(messages[-1].content): res = response diff --git a/playground/demos/chatbot/server/server_models.py b/playground/demos/chatbot/server/server_models.py index 49e54533..d7ade77c 100644 --- a/playground/demos/chatbot/server/server_models.py +++ b/playground/demos/chatbot/server/server_models.py @@ -44,8 +44,8 @@ class ChatRequest(BaseModel): """ The model for a ChatRequest for theChatbot Chat POST endpoint""" id: str model: AIModel = AIModel( - id="NousResearch/Meta-Llama-3-8B-Instruct", - name="NousResearch/Meta-Llama-3-8B-Instruct", + id="NousResearch/Meta-Llama-3.1-8B-Instruct", + name="NousResearch/Meta-Llama-3.1-8B-Instruct", maxLength=2048, tokenLimit=2048, ) diff --git a/playground/demos/vLLM/vLLM_example.py b/playground/demos/vLLM/vLLM_example.py index e4805838..090b902d 100644 --- a/playground/demos/vLLM/vLLM_example.py +++ b/playground/demos/vLLM/vLLM_example.py @@ -11,7 +11,7 @@ api_base = os.getenv("OPENAI_API_BASE") or "http://localhost:8000/v1" # for vllm # Create an instance of the OpenAIChat class model = OpenAIChat( - base_url=api_base, api_key=api_key, model="NousResearch/Meta-Llama-3-8B-Instruct", temperature=0.5, streaming=True, verbose=True + base_url=api_base, api_key=api_key, model="NousResearch/Meta-Llama-3.1-8B-Instruct", temperature=0.5, streaming=True, verbose=True ) # Initialize the agent