Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/phaniteja5789/langserve-ollama-codesummarization-using-docker
https://github.com/phaniteja5789/langserve-ollama-codesummarization-using-docker
docker docker-compose docker-image dockerfile langchain-python langserve ollama prompt python3
Last synced: 7 days ago
JSON representation
- Host: GitHub
- URL: https://github.com/phaniteja5789/langserve-ollama-codesummarization-using-docker
- Owner: phaniteja5789
- Created: 2024-05-11T16:02:52.000Z (6 months ago)
- Default Branch: main
- Last Pushed: 2024-05-11T17:12:16.000Z (6 months ago)
- Last Synced: 2024-05-12T17:51:44.441Z (6 months ago)
- Topics: docker, docker-compose, docker-image, dockerfile, langchain-python, langserve, ollama, prompt, python3
- Language: Python
- Homepage:
- Size: 409 KB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: Readme.md
Awesome Lists containing this project
README
This repository details out the code summarization using the Langchain framework with LangServe, Ollama and docker.
The sourcecode.py is the entry point for the application.
This project is developed in 2 phases
phase-1
The prompt has been developed using PromptTemplate
And using the PromptTemplate and the LLM a langchain Chain is created. And the chain will be invoked whenever the endpoint will be hit since its been integrated with LangServe
The Langserve internally uses the FastAPI uvicorn server and it will be running in the port 8080 of the Local host.
A Dockerfile is created in order to build a image, which uses the uvicorn server in order to run an API in the port 8080.
After completion of Phase-1, the image is built with the name langservecontainer.
Framework used in this Phase are langchain
Modules used in this Phase are
langchain_community.llms.ollama import Ollama ==> in order to use Ollama where the llama3 model will be running in the local host at port 11434
langserve import add_routes ==> used to run the Chain at the endpoint 8080.
langchain.prompts import PromptTemplate ==> used the PromptTemplate in order to create a Prompt that needs to be sent to the LLM.
phase-2
The DockerCompose.yaml is created, in this compose file, there are 2 images are used.
1st Image
In this Image, the Ollama image will be pulled from the repository and it will be running in the port 11434
The model that I am using is llama3 and it will be saved in the /root/.ollama
2nd Image
In this Image, the Langserve endpoint will be running which uses the Langchain Chain in order to generate the summary for the method and the result will be shown in the output of the LangServe Playground
The final output is shown as below
The prompt that is sent to the LLM(llama3) using Ollama in order to run the LLM locally is as below screenshot.
