CSV Analytics Workspace

CSV 3D Explorer

Upload a CSV, prepare the dataset, and inspect charts, statistics, distributions, and correlations in one focused workspace.

Data Preparation

Data prep overview

No dataset loaded

Review the prepared dataset at a glance, then open Data Prep to configure transformations and run diagnostics before analyzing the results.

Upload a CSV to start preparing the dataset.

No dashboard loaded yet

No Data Yet

Add data to generate a dashboard.

Drop in a CSV file and the interactive 3D view will appear here.

Data Prep

Workflow

Prepare the dataset, validate edge cases, and then analyze the prepared result.

Use transformations to shape the dataset and diagnostics to inspect unusual rows before moving back to the analysis tabs.

Pipeline Summary

Upload a CSV to begin preparing the dataset.

Draft changes are previewed here first and only affect the analytical tabs after you apply them.

Configure

Column transforms

Set numeric and categorical preparation steps before reviewing the output.

Scaling Normalize or log-transform numeric columns

Numeric Scaling

Apply a persistent scaling transform to selected numeric columns.

Log Scaling

Convert selected numeric columns to `log10(column)` before analysis.

When values are <= 0

Encoding Generate one-hot columns for categorical variables

Categorical Encoding

Create numeric one-hot columns from selected categorical variables.

Filter

Row selection rules

Control which rows are kept for analysis and which are explicitly removed.

Filtering Keep only rows that satisfy every subset rule

Subset Rows

Keep only rows that match every rule below.

No subset rules yet. Add a rule to keep only matching rows.

Exclusion Rules Remove rows that match any exclusion rule

Exclusion Rules

Exclude rows that match any rule below, similar to the existing filter workflow.

No exclusion rules yet. Add a rule to remove matching rows.

Preview

Prepared dataset

Review row counts, added columns, and sample values before applying the draft pipeline.

Prepared Dataset Preview

Preview how the applied pipeline will change rows, columns, and data types.

No preview available yet.

Outlier Review

Inspect potentially unusual rows in the prepared dataset using robust or standard numeric rules.

Method

Row Label

Sensitivity Threshold

1.50

Common IQR default is 1.5 x IQR.

Search numeric parameters

No numeric columns selected

Potential outliers are highlighted for review only. They may represent valid rare cases, not necessarily bad data.

Review Focus

Choose the histogram focus and scatter axes explicitly instead of relying on automatic picks.

Distribution Focus

Scatter X

Scatter Y

Column Summary

Ranked Rows

Scatter Review

Distribution Context

Distribution Explorer

Variable Selection

Numeric Variable

Split By

Primary Plot Settings

Primary View

Overlay

Histogram Settings

Histogram Scale

Histogram Bins

Histogram + KDE is the default overlay because it reveals both shape and smooth trend on the same axis.

Primary Distribution Plot

Shape Summaries

Box Plot

Violin Plot

CSV 3D Explorer

Data prep overview

Add data to generate a dashboard.

Loading dashboard...