mraniki/myllm

A python package to interact with llm model

👀 interested in algotrading,defi, continous integration and devops
📫 Reach out on

📈 More Stats

myllm's Issues

✨ Integrate huggingface_hub to download new model

Details

Documentation: https://hf.co/docs/huggingface_hub
Source Code: https://github.com/huggingface/huggingface_hub

1. login using settings.huggingface_api_token
   huggingface-cli login

or using an environment variable

huggingface-cli login --token $HUGGINGFACE_TOKEN

2. download and use a new model from a given repo provided as a string in the settings like settings.huggingface_ repo_id

from huggingface_hub import hf_hub_download
hf_hub_download(repo_id="tiiuae/falcon-7b-instruct", filename="config.json")

🚀 Integrate Bard Library

https://pypi.org/project/bardapi/

from bardapi import Bard

token = 'xxxxxxx'
bard = Bard(token=token)
bard.get_answer("나와 내 동년배들이 좋아하는 뉴진스에 대해서 알려줘")['content']

Dependency Dashboard

This issue lists Renovate updates and detected dependencies. Read the Dependency Dashboard docs to learn more.

Ignored or Blocked

These are blocked by an existing closed PR and will not be recreated unless you click a checkbox below.

• Update dependency g4f to v0.0.1.4
• Update dependency langchain to v0.0.256

Detected dependencies

dockerfile

.registry/Dockerfile

• python 3.11

github-actions

.github/workflows/👷Flow.yml

pip_requirements

.requirements/requirements.txt

• aiohttp ==3.8.5
• aiosignal ==1.3.1
• async-timeout ==4.0.2
• attrs ==23.1.0
• browser-cookie3 ==0.19.1
• certifi ==2023.7.22
• cffi ==1.15.1
• charset-normalizer ==3.2.0
• colorama ==0.4.6
• curl-cffi ==0.5.8.1
• dataclasses-json ==0.5.14
• dynaconf ==3.2.0
• frozenlist ==1.4.0
• g4f ==0.0.1.3
• greenlet ==2.0.2
• idna ==3.4
• jeepney ==0.8.0
• langchain ==0.0.255
• langsmith ==0.0.19
• loguru ==0.7.0
• lz4 ==4.3.2
• marshmallow-enum ==1.5.1
• marshmallow ==3.20.1
• multidict ==6.0.4
• mypy-extensions ==1.0.0
• numexpr ==2.8.5
• numpy ==1.25.2
• openapi-schema-pydantic ==1.2.4
• packaging ==23.1
• pycparser ==2.21
• pycryptodome ==3.18.0
• pycryptodomex ==3.18.0
• pydantic ==2.1.1
• pyexecjs ==1.5.1
• pyyaml ==6.0.1
• requests ==2.31.0
• six ==1.16.0
• sqlalchemy ==2.0.19
• tenacity ==8.2.2
• typing-extensions ==4.7.1
• typing-inspect ==0.9.0
• urllib3 ==2.0.4
• websockets ==11.0.3
• win32-setctime ==1.1.0
• yarl ==1.9.2

poetry

pyproject.toml

• dynaconf ^3.1.12
• loguru ^0.7.0
• g4f ^0.0.1
• python-semantic-release ^8.0.2
• ruff >=0.0.280,<0.0.283
• pytest ^7.0
• pytest-cov ^4.1
• pytest-asyncio ^0.21.0
• pytest-mock ^3.11.1
• pytest-loguru ^0.2.0
• sphinx ^7.0.0
• pydata-sphinx-theme ^0.13.3
• sphinx-hoverxref ^1.3.0
• sphinx-notfound-page ^0.8.3
• sphinx_copybutton 0.5.2
• myst_parser ^2.0.0
• sphinx_design ^0.5.0

---

• Check this box to trigger a request for Renovate to run again on this repository

Extend the talk method to support a discussion on the same topic

not the method receive one prompt and provide one response.
enhance to be able to keep the topic open.

🚀 Integrate openAI lib

https://github.com/openai/openai-python

from openai import AsyncOpenAI

client = AsyncOpenAI(
# This is the default and can be omitted
api_key=os.environ.get("OPENAI_API_KEY"),
)

async def main() -> None:
chat_completion = await client.chat.completions.create(
messages=[
{
"role": "user",
"content": "Say this is a test",
}
],
model="gpt-3.5-turbo",
)

asyncio.run(main())

✨ Integrate langchain package once it is pydantic v2 ready

Details

No response

🐛 MyLLM loop error to investigate

🚀 Sentimix: A Sentiment Analysis Tool for Mixed Texts

Given a specific text, analyze the sentiment of any text and provide a guidance based on the polarity and intensity of the emotions expressed. Sentimix can handle mixed texts, such as weather reports or market reports, that may contain both factual and subjective information. Sentimix can also identify the main topics and keywords of the text and provide a summary of the sentiment for each topic. Sentimix can help users to understand the tone and mood of the text and make informed decisions based on the guidance.

For a given text, provide Positive 🟦 / Negative 🟨 / Neutral 😐 guidance

Integrate Serp API in the llm talk method

Details

python to connect serpapi
from serpapi import GoogleSearch
search = GoogleSearch({
"q": "coffee",
"location": "Austin,Texas",
"api_key": ""
})
result = search.get_dict()

code snippet
from langchain import SelfAskWithSearchChain, OpenAI, SerpAPIChain

llm = OpenAI(temperature=0)
search = SerpAPIChain()

self_ask_with_search = SelfAskWithSearchChain(llm=llm, search_chain=search)

self_ask_with_search.run("What is the hometown of the reigning men's U.S. Open champion?")