Vanna Ask

First Post:

Last Update:

Vanna ASk

1
2
3
4
5
6
7
8
9
10
11
12
13
from src.vanna.ollama import Ollama
from src.vanna.chromadb import ChromaDB_VectorStore
from src.vanna.flask import VannaFlaskApp

class MyVanna(ChromaDB_VectorStore, Ollama):
    def __init__(self, config=None):
        ChromaDB_VectorStore.__init__(self, config=config)
        Ollama.__init__(self, config=config)

vn = MyVanna(config={'model': 'qwen2.5-coder:32b','ollama_host':'http://172.16.54.149:11434'})
vn.connect_to_mysql(user='root',password='123456',host='172.16.54.161',dbname='eims_test',port=3306)
VannaFlaskApp(vn, allow_llm_to_see_data=True).run(port=8085, host='0.0.0.0')

由于已经运行过 Vanna train 内对应的训练代码 上述不展示

接下来的问题测试均为 查询下 ys_vehicle_transfer 表中有哪些数据

运行

generate_sql

请求 URL:http://127.0.0.1:8085/api/v0/generate_sql
请求方法:GET

Method Target
Get http://127.0.0.1:8085/api/v0/generate_sql
Key Value
question “question”

return

1
2
3
4
5
{
  "id": "your id",
  "text": "SELECT * FROM ys_vehicle_transfer;",
  "type": "sql"
}

run_sql

请求 URL:http://127.0.0.1:8085/api/v0/run_sql
请求方法:GET

Method Target
Get http://127.0.0.1:8085/api/v0/run_sql
Key Value
id “id from generate_sql”

return

1
2
3
4
5
{
  "df": "df content",
  "id": "your id",
  "type": "df"
}

download_csv

请求 URL:http://127.0.0.1:8085/api/v0/download_csv
请求方法:GET

Method Target
Get http://127.0.0.1:8085/api/v0/download_csv
Key Value
id “id from generate_sql”

return

1
2
3
{
    csv表格信息
}

generate_plotly_code

请求 URL:http://127.0.0.1:8085/api/v0/generate_plotly_figure
请求方法:GET

Method Target
Get http://127.0.0.1:8085/api/v0/
Key Value
id “id from generate_sql”

return

1
2
3
4
5
{
  "fig": "{.........}",
  "id": "your id",
  "type": "plotly_figure"
}
≡