---
annotations_creators: []
language:
  - en
language_creators: []
license: []
multilinguality:
- monolingual
pretty_name: 'adaption-nl_to_viz_spec_pairs'
size_categories:
- 10K<n<100K
source_datasets:
- 'original'
tags:
  - adaption
  - instruction-tuning
  - data-analysis-visualization
task_categories: []
task_ids: []
---

![banner](https://proteus-prod-public.s3.us-east-1.amazonaws.com/temp/f6c5e1d2-b810-4926-93a5-b3758c769be7.png)

This dataset is a remastered version prepared using [Adaption's](https://adaptionlabs.ai/app/auth) Adaptive Data platform.

# adaption-nl_to_viz_spec_pairs

This dataset contains pairs of natural language requests and database schemas mapped to structured visualization specifications. Each entry includes a prompt with schema details and a user request, paired with a completion defining chart type, axes, grouping, and aggregation. The data is designed for training models to translate text queries into executable chart configurations.

### Dataset size

There are 11,698 data points in this dataset. This is an instruction tuning dataset.

### Quality of Remastered Dataset

The final quality is B, with a relative quality improvement of 5.0%.

### Domain
- Data-analysis-visualization (100%)

### Language
- English (100%)

### Tone
- Technical (100%)

### Evaluation Results

- **Quality Gains:**
<img src="https://proteus-prod-public.s3.us-east-1.amazonaws.com/temp/c2473e60-0e36-4801-8f4a-f919b4a56d04.png" alt="QualityGains" style="max-width: 50%; display: block; margin-left: auto; margin-right: auto;" />

- **Grade Improvement:**
<img src="https://proteus-prod-public.s3.us-east-1.amazonaws.com/temp/28a7c651-e209-4031-97c8-f4424e9981cf.png" alt="Grade" style="max-width: 50%; display: block; margin-left: auto; margin-right: auto;" />

- **Percentile Chart:**
<img src="https://proteus-prod-public.s3.us-east-1.amazonaws.com/temp/6edea423-4dbb-445b-bbba-880a3114706a.png" alt="Percentile Chart" style="max-width: 50%; display: block; margin-left: auto; margin-right: auto;" />