LLM讀書會 - 財報機器人專案

主辧單位: 台灣人工智慧協會進修交流委員會、H.I.T. 生醫創新實作社群

This is a side project from one of the TAIA study groups. The purpose is to build a chatbot and interact with users to analyse and build a profitable investment for seeking alpha.

Important dependency docs/repo

Langchain: Doc | Repo Chainlit: Doc | Repo

Preparation / Installation environment

• English

1. Install Anaconda or Miniconda.

2. Install Git

3. Change dir to the path you will place the project. Dowland or clone this repo

    git clone https://github.com/email81227/llm-study.git
   

4. Build the project enviroment

    conda create -n llm-study python=3.10 poetry
   

5. Activate the env. (To deactivate, using conda deactivate)

    conda activate llm-study 
   

6. Switch to the project dir

    cd {path to the repo cloned}/llm-study
   

7. Install packages by poetry

    poetry install
   

8. Create .env file under the project dir

9. Open .env and append the following line

    OPENAI_API_KEY="{your openai key}"
    USER_EMAIL="{your email address}"
    USER_NAME="{your nickname}"
    LANGFUSE_PUBLIC_KEY="{apply here: https://cloud.langfuse.com/}"
    LANGFUSE_SECRET_KEY="{apply here: https://cloud.langfuse.com/}"
   

• 中文說明

1. 安裝 Anaconda 或 Miniconda

2. 安裝 Git

3. 移動到預計放置專案的目錄下，下載或複製專案

    git clone https://github.com/email81227/llm-study.git
   

4. 建立Python專案環境

    conda create -n llm-study python=3.10 poetry
   

5. 切換到該環境（用conda deactivate反過來切換回base env）

    conda activate llm-study 
   

6. 切換到專案目錄下

    cd {path to the repo cloned}
   

7. 安裝專案內套件

    poetry install 
   

8. 新增名為.env的空白檔案

9. 打開或新增以下內容至.env中

    OPENAI_API_KEY="{your openai key}"
    USER_EMAIL="{your email address}"
    USER_NAME="{your nickname}"
    LANGFUSE_PUBLIC_KEY="{apply here: https://cloud.langfuse.com/}"
    LANGFUSE_SECRET_KEY="{apply here: https://cloud.langfuse.com/}"
   

Execution

Run the command on project dir:

chainlit run chat.py -w