NickyNicky/synthetic_text_to_sql_format_chatML_gemma

---
dataset_info:
features:
- name: Text
dtype: string
- name: len_token_text
dtype: int64
splits:
- name: train
num_bytes: 142483016
num_examples: 100000
- name: test
num_bytes: 8325593
num_examples: 5851
download_size: 44871261
dataset_size: 150808609
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
license: apache-2.0
language:
- en
---

<p align="center">
<img src="https://cdn-uploads.huggingface.co/production/uploads/641b435ba5f876fe30c5ae0a/bXI58O_0PXDmFGl8EeK7F.png" alt="" style="width: 60%; max-height: 550px;">
</p>

## dataset base.
```
gretelai/synthetic_text_to_sql
```

```py
dataset = load_dataset("NickyNicky/synthetic_text_to_sql_format_chatML_gemma")
```

```
<bos><start_of_turn>system
You are a helpful AI assistant.
you are a sql expert who responds in json format.<end_of_turn>
<start_of_turn>user
## prompt:
What is the total gold production by 'Site B' in the 'production' table?

## sql context:
CREATE TABLE production (id INT, site VARCHAR(50), year INT, gold_production INT, silver_production INT);<end_of_turn>
<start_of_turn>model
{
"sql_explanation": "The SQL query calculates the total gold production for 'Site B' by using the SUM function on the 'gold_production' column in the 'production' table, filtered by the 'site' column equal to 'Site B'.",
"sql_complexity_description": "basic SQL with a simple select statement",
"sql_complexity": "basic SQL",
"sql": "SELECT SUM(gold_production) FROM production WHERE site = 'Site B';",
"domain_description": "Mining operation data, environmental impact stats, workforce diversity, and resource depletion metrics.",
"sql_task_type_description": "generating reports, dashboards, and analytical insights",
"sql_task_type": "analytics and reporting",
"domain": "mining industry"
}<end_of_turn><eos>
```

```
<bos><start_of_turn>system
You are a helpful AI assistant.
you are a sql expert who responds in json format.<end_of_turn>
<start_of_turn>user
## prompt:
What is the total number of ground vehicles sold by GlobalTechDefense in the Middle East?

## sql context:
CREATE TABLE IF NOT EXISTS sales (equipment_type varchar(20), quantity_sold int, company varchar(30), region varchar(30));
INSERT INTO sales (equipment_type, quantity_sold, company, region) VALUES ('tank', 150, 'GlobalTechDefense', 'Middle East'), ('jeep', 75, 'GlobalTechDefense', 'Middle East'), ('truck', 120, 'GlobalTechDefense', 'Middle East');<end_of_turn>
<start_of_turn>model
{
"sql_explanation": "The SQL query calculates the total number of ground vehicles sold by GlobalTechDefense in the Middle East by summing the quantity_sold column. It filters for the relevant equipment_type, company, and region.",
"sql_complexity_description": "basic SQL with a simple select statement",
"sql_complexity": "basic SQL",
"sql": "SELECT SUM(quantity_sold) FROM sales WHERE equipment_type = 'ground vehicles' AND company = 'GlobalTechDefense' AND region = 'Middle East';",
"domain_description": "Military equipment sales data, defense project timelines, contract negotiations, and geopolitical risk assessments.",
"sql_task_type_description": "generating reports, dashboards, and analytical insights",
"sql_task_type": "analytics and reporting",
"domain": "defense contractors"
}<end_of_turn><eos>
```

<!-- https://colab.research.google.com/drive/11nF88nFHe-zHNxZf-aFpwASj6Txe979c#scrollTo=tF-dM-wFjDJN -->

原始信息汇总

数据集概述

数据集特征

• Text: 数据类型为字符串（string）。
• len_token_text: 数据类型为整数（int64）。

数据集拆分

• 训练集（train）: 包含100000个示例，总大小为142483016字节。
• 测试集（test）: 包含5851个示例，总大小为8325593字节。

数据集大小

• 下载大小: 44871261字节。
• 数据集总大小: 150808609字节。

配置

• 默认配置（default）:
  • 训练集路径: data/train-*
  • 测试集路径: data/test-*

许可证

• 许可证: Apache-2.0

语言

• 语言: 英语（en）