Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Data-Copilot: Bridging Billions of Data and Humans with Autonomous Workflow
| Date | Stars |
|---|---|
| 2026-07-31 | 1512 |
| 2026-08-03 | 1512 |
| 2026-08-06 | 1512 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
# Data-Copilot
[](https://huggingface.co/spaces/zwq2018/Data-Copilot)
[](https://arxiv.org/abs/2306.07209)
[](https://zhuanlan.zhihu.com/p/636906119)
## Overview
Data-Copilot is a LLM-based system that help you address data-related tasks.
Data-Copilot connects data sources from different domains and diverse user tastes, with the ability to autonomously manage, process, analyze, predict, and visualize data.
<img src="./assets/Word Art.png" alt="Image" style="width: 900px;">
See our paper: [Data-Copilot: Bridging Billions of Data and Humans with Autonomous Workflow](https://arxiv.org/abs/2306.07209), Wenqi Zhang, Yongliang Shen, Weiming Lu, Yueting Zhuang
## 🔥Demo [video](https://zhuanlan.zhihu.com/p/636906119)
Since gpt3.5 has only a 4k input token limit, it currently can access to Chinese stocks, funds and some economic data.
**The Data-Copilot can query and predict data autonomously**:
<img src="./demo1.png" alt="Image" style="width: 900px;">
Support model and data sources:
| | CHN Stock | CHN Fund |CHN Economic data | CHN Financial data |
|:-------------:|:---------:|:--------:|:---------------:|:------------------------:|
| Openai-GPT3.5 | ✓ | ✓ | ✓ | ✓ |
| Azure-GPT3.5 | ✓ | ✓ | ✓ | ✓ |
| Qwen-72b-Chat | ✓ | ✓ | ✓ | ✓ |
<img src="./assets/fig1.png" alt="Image" style="width: 900px;">
We propose Data-Copilot, an LLM-based system linking Chinese financial markets such as stock, funds, economic, financial data, and live news
* ⭐ Data-Copilot can **autonomously manage, process, analyze, predict, and visualize data**. When a request is received, it transforms raw data into informative results that best match the user’s intent.
* ⭐ Acting as a **designer**: Data-Copilot independently designs versatile interface tools with different functions through self-request and iterative refinement.
* ⭐ As a **dispatcher**: DataCopilot adeptly invokes the corresponding interfaces sequentially or in parallel and transforms raw data from heterogeneous sources into graphics, tables, and text, without human assistance.
## 🌳 QuickStart
First replace openai.key and Tushare token in main.py with your personal Openai key and [Tushare token](https://tushare.pro/).
The organization of the whole project is as follows:
```angular2html
|-- README.md
|-- app.py
|-- create_tool
| |-- Atomic_api_json.py
| `-- all_atomic_api.json
|-- lab_gpt4_call.py
|-- lab_llms_call.py
|-- main.py
|-- output
|-- prompt_lib
| |-- prompt_economic.json
| |-- prompt_financial.json
| |-- prompt_fund.json
| |-- prompt_intent_detection.json
| |-- prompt_stock.json
| |-- prompt_task.json
| `-- prompt_visualization.json
|-- requirements.txt
|-- tool.py
|-- tool_lib
| |-- atomic_api.json
| |-- tool_backup.json
| |-- tool_economic.json
| |-- tool_financial.json
| |-- tool_fund.json
| |-- tool_stock.json
| `-- tool_visualization.json
```
app.py is the file to start gradio. main.py is the processing flow of interface scheduling, and lab_gpt4_call.py is the file to call the GPT35 model.
The tool_lib and tool.py contain the interface tools obtained after the first phase of interface design.
The folder prompt_lib contains the design of the prompt and the in context demonstration.
### Requirements
```bash
pip install -r requirements.txt
```
Then run the following command:
### For Local
```bash
python main.py
```
You can select the LLM in main.py by setting:
```
model='<the model you choose>'
```
Remember to fill in the key of the LLM you chose:
- For GPT, fill in the key of Openai in `main.py`
```
openai_key = os.getenv("OPENAI_KEY")
```
- For Qwen-72b-Chat, Excerpt of 6,353 characters
Read on GitHub
Wenqi Zhang · Zhejiang University · China
26
6
1
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:6db1ba04b22fd8e3, llm:Repository title and description: 'Data-Copilot: Bridging Billions of Data and Humans with Autonomous Workflow' (Python). Suggests autonomous data workflows/agents working with large datasets—likely an agent/RAG or data-processing platform for AI assistants.
matched fp:6db1ba04b22fd8e3, llm:Repository title and description: 'Data-Copilot: Bridging Billions of Data and Humans with Autonomous Workflow' (Python). Suggests autonomous data workflows/agents working with large datasets—likely an agent/RAG or data-processing platform for AI assistants.
matched fp:6db1ba04b22fd8e3, llm:Repository title and description: 'Data-Copilot: Bridging Billions of Data and Humans with Autonomous Workflow' (Python). Suggests autonomous data workflows/agents working with large datasets—likely an agent/RAG or data-processing platform for AI assistants.