Explore results in one workspace

Keep measurement tables, descriptive summaries and plots beside their controls in Results Workspace, without replacing your workflow nodes.

New in 0.16.0a1

Results Workspace brings existing table, Statistics and Plot Results tools into one window. It does not add statistical tests or change their calculation methods.

Open the workspace

1. Calculate a table-producing node, such as Measure Objects, Merge Tables or Table Source.
2. Select it and choose Open Results Workspace… in the inspector. You can also open the workspace from Statistics or Plot Results.
3. Use the selection bar above the tabs: Workflow → Data source → Statistics node → Plot. Matching blue arrows join all four selectors. Bold headings and a subtle border highlight this global connection bar, which stays visible in every view. All four dropdowns share one aligned row, with + beside Statistics node and Plot to add a connected summary or plot.
4. Choose Data, Summary or Plots using the full-width icon tabs. Tabs change the view, not the selected source, summary or plot. Controls are on the left and the relevant table or figure is on the right.

Use the window's title-bar Maximize button for more room. Restore returns it to its previous size; both actions leave the analysis unchanged.

New in 0.16.0a4: independent results window

Use the native Minimize control, or double-click the title bar or unused space in the top selection toolbar to maximize or restore. On Windows, Results Workspace can stay on another monitor while napari or detached VIPP is minimized. Reopening the minimized workspace restores it and retains its previous maximized state. Ending the host session closes the window; see Batch and Results windows for the platform boundary.

Choose the chain you want to explore:

• Workflow lists every open workflow. Selecting one activates its workflow tab and lists its table sources. Repeated workflow names include their tab number so you can tell them apart. A workflow with no table-producing nodes remains selectable and explains why no data sources are available.
• Data source identifies an exact table output, including its named port when relevant. Filtered and merged tables remain separate choices from their upstream measurements.
• Statistics node selects a summary connected to that table. Choose None — use input data to work directly with the selected table instead.
• Plot lists only plots directly connected to the selected Statistics output, or to the data source when Statistics is None. If none exist, VIPP shows No connected plots, not a plot from another branch.

The relationship caption confirms the selected chain. These selections change what you browse; they do not edit graph connections. Settings changes, + actions and calculations apply to the selected workflow. The editing heading identifies the node for the current tab, and Show node locates it in the workflow.

Selectors use descriptive node names instead of relying on numbered labels such as Statistics 1. Table Source uses the dataset title or filename; Statistics uses the measurements and grouping, adding Image averages or Sample averages for those observation levels; Plot Results uses a meaningful figure title or its plot settings. For example, a summary might read 3 measurements by Well, while a plot might read area distribution. These examples depend on the actual selected columns.

Operation type and current settings remain available alongside the name or in its tooltip. If names match, VIPP adds source context and, when necessary, a short stable identifier. Use Show node, then the inspector's Name field or the node's Rename… action to give it your own description. See Name a node for automatic naming, reset and save behavior. A custom plot-node name is independent of the exported figure's title.

For multiple sources, first merge matching measurements or collect batch measurements. Merging adds fields for matching objects; collection brings observations from several images into one dataset. Browsing another workflow does not copy its data, combine tables or create connections between workflows.

Review the data

In Data, search rows and choose which columns to show. These controls only change what you see: they do not exclude observations from a summary or plot.

New in 0.16.0a4: paired column selection

Use Select all or Deselect all above Visible columns to show or hide all columns at once, then check the individual columns you want. The scope is all columns; row search does not change that scope.

Export uses the complete table, not just the visible search results or columns. For a saved column selection used downstream, use Select Table Columns in the workflow. Its paired selection controls also affect all columns, and its choices change the downstream table.

Increase Decimal and Decrease Decimal controls change how many decimal places the Data, Summary and View plotted data tables display, starting at 3. Hover over a cell for its full original value. Each table view keeps its own choice across refreshes; analysis and exports remain unchanged. See Decimal places in table views for limits and precision details.

Build a summary

Select a Statistics node in the top bar, or use + beside it to add an ordinary Statistics node connected to the selected data source. Then open Summary. Choose measurements, groups, observation level and statistics on the left; review the updated table and Result overview on the right. The overview separates objects used/excluded from image or sample averages summarized, for each measurement across all groups. Enable Show all fields to see per-group inclusion counts and calculation settings; table export includes them whether or not they are currently visible.

Object values, image averages and sample averages have different meanings. See Summarize measurements for identity fields, weighting and missing-value choices. Counts and units stay with the result.

Choose what the plot uses

Choose Plot in the top bar, then open Plots. The + beside Plot creates a normal Plot Results node connected to the selected Statistics output, or to Data source when Statistics is None — use input data. Opening a tab never creates a node or selects a different branch.

To deliberately reconnect an existing plot, expand Change plot input… in the plot controls and choose its new input:

Source	What the connected rows represent
Original measurements	Rows from the workspace's source table. For an object table, these are individual objects; Mean per image can explicitly reduce them to image means.
Summary table	Rows already calculated by the selected Statistics node, often one row per group. Choose a summary column such as a mean to plot it.

The blue-bordered input card stays visible above the figure. Its header identifies the input; click the header or the outlined cyan chevron at its right edge to expand Input details. The body starts collapsed and explains the source, what each plotted value represents and the counts. Original measurements means the workspace's connected source, not earlier raw rows. When that source is already a summary, the input selector calls it Input summary table.

Choosing a summary in Change plot input connects the plot to that Statistics node and uses each summary row directly, rather than averaging it again by image. Review the measurement and group fields after changing sources: a column available in the original table might not exist in the summary. VIPP does not guess an equivalent measurement.

For a new plot of a summary table, first choose Measurement (and both axes for Scatter). VIPP waits for these choices before calculating the plot; the setup prompt is not a failed analysis. It never picks a summary statistic on your behalf. Once configured, changes update the plot automatically.

A summary does not contain the original observations

Two treatment means are two summary values, not all the original objects or independent samples. Selecting a mean column does not automatically add its SD as error bars or recover the underlying distribution.

The plotting controls are the same as Plot Results, including Appearance and automatic/custom numeric-axis intervals.

Messages beside the plot distinguish guidance from problems:

• Grey analysis notes explain interpretation: summary rows are not original objects, and SD is not automatically an error bar.
• Orange cautions flag issues to review, such as excluded rows, values excluded by log scales, display sampling, many numeric categories or no eligible values.
• Red errors mean the plot could not be calculated. They remain visible until the problem is addressed.

Analysis notes appear by default. Dismiss them after reading, and use the compact Analysis notes indicator to reopen them. Notes return when input data, analysis choices (including grouping, averaging or log scales), or warning text changes. Switching tabs, resizing and appearance-only edits do not reopen them. Dismissal applies only to that plot in this workspace window; calculations, exports, inspector and separate plot-window warnings are unchanged. Errors, setup prompts and out-of-date status remain visible and cannot be dismissed this way.

The figure fits the available preview area, including its axes. Maximize or resize the window, or drag the divider to give it more room; the settings scroll independently. Resizing the preview does not change the measurements, saved plot recipe or export dimensions. Set the exported figure's width and height in Export figure….

Keep, export and close

• The + actions add normal Statistics or Plot Results nodes. Opening the workspace or switching its tabs does not create nodes.
• Changes edit the selected node immediately and are shared with its inspector and existing plot window. Show node locates it in the graph. It hides (rather than closes) the workspace and centers the node; reopening Results Workspace from the inspector retains your view and selections. It targets the exact node shown in the editing heading and is disabled when there is no selected node, such as a summary with no connected plot. Different summary/plot nodes keep independent settings.
• Save each edited workflow to keep its nodes, connections and settings. Closing the workspace does not discard edits.
• Table export saves the complete current table as CSV/TSV, including summary counts and method fields. Excel export remains available in the separate batch collection window.
• Figure export uses Export figure… for sized PNG/TIFF/SVG/PDF output. A stale or calculating result is not exportable as a current result.

Changing results settings reuses current upstream measurements. If the measurement table itself is stale, recalculate that source before treating the summary or plot as current. The activity area remains in place while updates are prepared, so the controls do not jump.