Violin

Overview

The violin insight type is used to create violin plots, visualizing the distribution of numerical data. Violin insights combine aspects of box plots and density plots to show the shape of the data distribution, including probability density. They are ideal for comparing distributions across categories.

You can customize the orientation, box overlay, points, and colors to represent your distribution data effectively.

Common Uses

• Distribution Analysis: Visualizing the distribution of numerical data, similar to box plots but with density information.
• Comparing Categories: Comparing distributions across multiple categories.
• Outlier Detection: Identifying outliers and distribution shape.

Check out the Attributes for the full set of configuration options

Examples

Common Configurations

Simple Violin InsightViolin Insight with Box OverlayViolin Insight with Split Categories

sources:
  - name: violin-data-source
    type: duckdb
    database: target/seeds/violin_data.duckdb
    seeds:
      - table_name: model
        args:
          - echo
          - |
            category,value
            A,23
            A,25
            A,27
            B,30
            B,35
            B,28
models:
  - name: violin-data
    source: ${ref(violin-data-source)}
    sql: select * from model
insights:
  - name: Simple Violin Insight
    props:
      type: violin
      x: ?{${ref(violin-data).category}}
      y: ?{${ref(violin-data).value}}
charts:
  - name: Simple Violin Chart
    insights:
      - ${ref(Simple Violin Insight)}
    layout:
      title:
        text: Simple Violin Plot<br><sub>Distribution of Values by Category</sub>

sources:
  - name: violin-data-box-source
    type: duckdb
    database: target/seeds/violin_data_box.duckdb
    seeds:
      - table_name: model
        args:
          - echo
          - |
            category,value
            A,23
            A,25
            A,27
            B,30
            B,35
            B,28
models:
  - name: violin-data-box
    source: ${ref(violin-data-box-source)}
    sql: select * from model
insights:
  - name: Violin Insight with Box
    props:
      type: violin
      x: ?{${ref(violin-data-box).value}}
      y: ?{${ref(violin-data-box).category}}
      orientation: h
      box:
        visible: true
      points: "all"
      marker:
        symbol: "cross-dot"
charts:
  - name: Violin Chart with Box
    insights:
      - ${ref(Violin Insight with Box)}
    layout:
      title:
        text: Violin Plot with Box Overlay<br><sub>Distribution with Box Plot and Data Points</sub>

sources:
  - name: violin-data-split-source
    type: duckdb
    database: target/seeds/violin_data_split.duckdb
    seeds:
      - table_name: model
        args:
          - echo
          - |
            category,sub_category,value
            A,X,23
            A,Y,25
            A,Y,70
            A,Y,15
            A,X,27
            A,X,13
            A,X,21
            A,X,81
            B,X,30
            B,X,35
            B,X,4
            B,Y,35
            B,Y,6
            B,Y,5
            B,X,28
models:
  - name: violin-data-split
    source: ${ref(violin-data-split-source)}
    sql: select * from model
insights:
  - name: Violin Insight Category X
    props:
      type: violin
      x: ?{${ref(violin-data-split).category}}
      y: ?{${ref(violin-data-split).value}}
      side: positive
    interactions:
      - filter: ?{${ref(violin-data-split).sub_category} = 'X'}
      - split: ?{${ref(violin-data-split).sub_category}}

  - name: Violin Insight Category Y
    props:
      type: violin
      x: ?{${ref(violin-data-split).category}}
      y: ?{${ref(violin-data-split).value}}
      side: negative
    interactions:
      - filter: ?{${ref(violin-data-split).sub_category} = 'Y'}
      - split: ?{${ref(violin-data-split).sub_category}}

charts:
  - name: Violin Chart with Split Categories
    insights:
      - ${ref(Violin Insight Category Y)}
      - ${ref(Violin Insight Category X)}
    layout:
      title:
        text: Violin Plot with Split Categories<br><sub>Side-by-Side Comparison of Distributions</sub>