Skip to main content

一个AI工具包,帮助用户快速调用小思框架(Xiaothink)相关接口。

Project description

Xiaothink Python 模块使用文档

Xiaothink 是一个以自然语言处理(NLP)为核心的AI研究组织,致力于提供高效、灵活的工具来满足各种应用场景的需求。Xiaothink Python 模块是该组织提供的核心工具包,涵盖了图像生成、文本续写、颜值评分以及对话模型等多种功能。以下是详细的使用指南和代码示例。

目录

  1. 安装
  2. 本地对话模型

安装

首先,您需要通过 pip 安装 Xiaothink 模块:

pip install xiaothink

注:由于业务范围调整,2025年7月17日后,小思框架将暂停所有WebAI服务,转向端侧AI模型领域研究,本代码库也同步删除相关接口。

本地对话模型

对于本地加载的对话模型,根据模型类型的不同,应调用相应的函数来进行对话。

单轮对话

适用于单轮对话场景。

示例代码

import xiaothink.llm.inference.test_formal as tf

model = tf.QianyanModel(
    ckpt_dir=r'path/to/your/t6_model',
    MT='t6_beta_dense',
    vocab=r'path/to/your/vocab'# vocab文件在模型储存库中已给出
)

while True:
    inp = input('【问】:')
    if inp == '[CLEAN]':
        print('【清空上下文】\n\n')
        model.clean_his()
        continue
    re = model.chat_SingleTurn(inp, temp=0.32)  # 使用 chat_SingleTurn 进行单轮对话
    print('\n【答】:', re, '\n')

多轮对话

适用于多轮对话场景。

示例代码

import xiaothink.llm.inference.test_formal as tf

model = tf.QianyanModel(
    ckpt_dir=r'path/to/your/t6_model',
    MT='t6_beta_dense',
    vocab=r'path/to/your/vocab'# vocab文件在模型储存库中已给出
)

while True:
    inp = input('【问】:')
    if inp == '[CLEAN]':
        print('【清空上下文】\n\n')
        model.clean_his()
        continue
    re = model.chat(inp, temp=0.32)  # 使用 chat 进行单轮对话
    print('\n【答】:', re, '\n')

文本续写

适用于更灵活的文本续写场景

示例代码

import xiaothink.llm.inference.test as test

MT = 't6_beta_dense'
m, d = test.load(
    ckpt_dir=r'path/to/your/t6_model',
    MT='t6_beta_dense',
    vocab=r'path/to/your/vocab'# vocab文件在模型储存库中已给出
)





inp='你好!'
belle_chat = '{"conversations": [{"role": "user", "content": {inp}}, {"role": "assistant", "content": "'.replace('{inp}', inp)    # t6系列中经过指令微调的模型支持的instruct格式
inp_m = belle_chat

ret = test.generate_texts_loop(m, d, inp_m,    
                               num_generate=100,
                               every=lambda a: print(a, end='', flush=True),
                               temperature=0.32,
                               pass_char=['▩'])    #▩是t6系列模型的<unk>标识

重要提示:对于本地模型,单论对话模型应调用 model.chat_SingleTurn 函数,多轮对话模型应调用 model.chat 函数,未进行指令微调的预训练模型模型建议调用 model.chat 函数。


以上就是 Xiaothink Python 模块的主要功能及使用方法。

如有任何疑问或建议,请随时联系我们:xiaothink@foxmail.com

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

xiaothink-1.0.5.tar.gz (25.0 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

xiaothink-1.0.5-py3-none-any.whl (26.5 kB view details)

Uploaded Python 3

File details

Details for the file xiaothink-1.0.5.tar.gz.

File metadata

  • Download URL: xiaothink-1.0.5.tar.gz
  • Upload date:
  • Size: 25.0 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.0.1 CPython/3.8.10

File hashes

Hashes for xiaothink-1.0.5.tar.gz
Algorithm Hash digest
SHA256 31a7b971c817011e55466f1920587410bb5df32c956e8984ada244e0566abd96
MD5 83de63d945a1077162d8b19512c4d42d
BLAKE2b-256 5749a6d186033530b1e42eed3ed6431421c0b0b2af97ace5e2650198b9b95049

See more details on using hashes here.

File details

Details for the file xiaothink-1.0.5-py3-none-any.whl.

File metadata

  • Download URL: xiaothink-1.0.5-py3-none-any.whl
  • Upload date:
  • Size: 26.5 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.0.1 CPython/3.8.10

File hashes

Hashes for xiaothink-1.0.5-py3-none-any.whl
Algorithm Hash digest
SHA256 7fa86791236af3bac5088f59fff33806f9a3592956d9105ad2a99a7c1edc17d3
MD5 7da0616d711f39958d3d827165e0c464
BLAKE2b-256 51c443675713c583f9b52d7b919c5c3bad5fd8d92b529d3e6b6c1fc6e8807df8

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page