Box Plot Calculator for Machine Learning Data

Calculated Results Waiting for data

Statistical Summary

Dataset	n	Min	Q1	Median	Q3	Max	IQR	Mean	SD	Outliers

Interpretation

Detected Outliers

Processed Data Table

#	Dataset	Value	Status	Z-score	Modified Z-score

1. Data Input

Machine Learning Preset

CSV Layout

Upload CSV

Datasets

Use one dataset per line. Start each line with a name followed by a colon.

Editable Values

Dataset	Value	Action

2. Statistical Methods

Quartile Method

Outlier Detection

IQR Multiplier

Z-score Threshold

Modified Z-score Threshold

Confidence Level

Bootstrap Repetitions

Missing Values

Decimal Precision

4. Chart Customization

Chart Title

Subtitle

Numeric Axis Title

Value Suffix

Box Width

Point Size

Point Opacity

Jitter Amount

Base Font Size

Chart Theme

Mean Marker

Export Background

5. Axis Controls

Numeric Scale

Custom Minimum

Custom Maximum

Tick Interval

Category Label Angle

Reverse numeric axis

6. Calculate and Manage

Import Configuration

Formula Used

Interquartile range: IQR = Q3 − Q1.

Standard Tukey fences: Lower fence = Q1 − 1.5 × IQR. Upper fence = Q3 + 1.5 × IQR.

Z-score: z = (x − mean) ÷ standard deviation.

Modified Z-score: 0.6745 × (x − median) ÷ MAD.

Approximate notched interval: median ± 1.57 × IQR ÷ √n.

How to Use

Enter one named dataset on each line.
Select quartile, outlier, and whisker methods.
Choose the chart style and display controls.
Press Calculate Box Plot to view results.
Export the chart, summary, outliers, or configuration.

Example Data

Dataset	Values	Suggested Use
Model A	0.81, 0.83, 0.79, 0.85	Cross-validation accuracy
Model B	0.76, 0.80, 0.78, 0.82	Model comparison
Latency	12, 13, 11, 14, 32	Outlier detection

Machine Learning Applications

Model EvaluationCompare fold scores, residuals, errors, and calibration metrics.

Feature AnalysisInspect feature spread, skew, instability, and unusual observations.

Performance TestingCompare training time, latency, throughput, and resource measurements.

Data QualityFind outliers, inconsistent groups, distribution shifts, and suspicious values.

Frequently Asked Questions

What does the box represent?

The box spans Q1 through Q3. It contains the middle fifty percent.

What does the median line show?

The line marks the middle ordered value. Half the observations lie on either side.

How are outliers detected?

The default method uses Tukey fences. Custom IQR and Z-score methods are available.

Can I compare multiple models?

Yes. Enter one line for each model or upload separate CSV columns.

Why do quartile methods differ?

Software packages use different interpolation rules. Small datasets show the largest differences.

What is a notched box plot?

A notch estimates median uncertainty. Non-overlapping notches can suggest different medians.

Can whiskers show minimum and maximum?

Yes. Select minimum to maximum from the whisker menu.

Does the calculator support negative values?

Yes. Decimal, negative, repeated, and scientific-notation values are accepted.

Can results be exported?

Yes. Download images, PDF reports, summary CSV files, and outlier reports.