Skip to main content
Open In ColabOpen on GitHub

Pipeshift

这将帮助您开始使用 LangChain 的 Pipeshift 完成模型（LLM）。有关 Pipeshift 功能和配置选项的详细文档，请参阅API 参考。

概述​

集成详情​

类包本地可序列化JS 支持包下载量包最新版本
Pipeshiftlangchain-pipeshift❌-❌PyPI - DownloadsPyPI - Version

设置​

要访问 Pipeshift 模型，您需要创建一个 Pipeshift 账户，获取一个 API 密钥，并安装 langchain-pipeshift 集成包。

凭证​

前往 Pipeshift 注册 Pipeshift 并生成 API 密钥。完成此操作后，请设置 PIPESHIFT_API_KEY 环境变量：

import getpass
import os

if not os.getenv("PIPESHIFT_API_KEY"):
    os.environ["PIPESHIFT_API_KEY"] = getpass.getpass("Enter your Pipeshift API key: ")

如果你想获取模型调用的自动化追踪，你也可以通过取消下面这行代码的注释来设置你的 LangSmith API 密钥：

# os.environ["LANGSMITH_TRACING"] = "true"
# os.environ["LANGSMITH_API_KEY"] = getpass.getpass("Enter your LangSmith API key: ")

安装​

LangChain Pipeshift 集成位于 langchain-pipeshift 包中：

%pip install -qU langchain-pipeshift
Note: you may need to restart the kernel to use updated packages.

实例化​

现在我们可以实例化我们的模型对象并生成聊天补全了：

from langchain_pipeshift import Pipeshift

llm = Pipeshift(
    model="meta-llama/Meta-Llama-3.1-8B-Instruct",
    temperature=0,
    max_tokens=512,
)

调用​

input_text = "Pipeshift is an AI company that "

completion = llm.invoke(input_text)
completion
'\xa0specializes in digital transformation and innovation for industry leaders. We leverage AI and IoT technology to provide data-driven insights, predictive analysis and decision support systems to our clients.\nAbout our company:\nOur company is built around the passion of creating positive impact through AI innovation. Pipeshift brings together top talent in AI, data science, software engineering, and business consultancy to deliver tailored solutions that drive growth and improvement for our clients across various industries.\nOur Mission and Values:\nOur mission is to revolutionize industries by harnessing the power of emerging technologies, like AI and IoT, to unlock new potential and drive progress. Our values are built around collaboration, innovation, integrity, customer value, and continuous learning.\nJob Title: Software Engineer (Mobile App Development)\nAbout the role:\nAs a Software Engineer (Mobile App Development) at Pipeshift, you will be at the forefront of cutting-edge mobile app development. Your responsibilities will include designing, developing, and implementing scalable and reliable mobile apps for various platforms, such as Android and iOS.\nResponsibilities\nDesign and develop mobile apps for various platforms using Java, Kotlin, Swift, or React Native\nWork collaboratively with cross-functional teams to gather requirements, plan and prioritize projects\nImplement UI/UX design principles to deliver user-friendly and visually appealing apps\nWrite clean, maintainable, and efficient code, adhering to best coding practices and coding standards\nTest and debug mobile apps to ensure high-quality, reliable, and stable performance\nCollaborate with the QA team to ensure all mobile apps meet or exceed quality and performance expectations\nStay up-to-date with industry trends, new technologies, and platform updates to enhance app development skills\nParticipate in code reviews to ensure code quality and adherence to coding standards\nSupport the DevOps team in continuous integration and delivery of mobile apps\nParticipate in the design and maintenance of technical documentation, coding standards and guidelines\nConduct peer mentoring and training to promote knowledge sharing and growth\nExperience and requirements\n2+ years of experience in mobile app development, or relevant technology experience with strong passion for mobile development.\nProficient in one or more mobile app development frameworks: React Native, iOS (Swift), Android (Kotlin/Java), Flutter, etc.\nStrong understanding of OOP (object-oriented programming) principles, design patterns, and data structures.\nExperience with Agile methodologies, version control systems (e.g., Git), and CI/CD pipelines (e.g., Jenkins, Travis CI, CircleCI).\nGood coding skills in languages such as Java, Kotlin, Swift,'

Chaining​

我们也可以将我们的 llm 与一个 prompt template 进行链接

API 参考​

有关 Pipeshift 所有功能和配置的详细文档，请前往 API 参考：https://dashboard.pipeshift.com/docs