Skip to content

Outputs

An Output block saves a pipeline result back into FlaskTrack.

Use it when the result should remain available after the run finishes.

Example

A pipeline might produce:

Python:normalized
Output

Configure the Output block:

Input:
input

Type:
Arrow table

Destination path:
analysis/plate-reader

Filename:
normalized-results.arrow

When the pipeline runs successfully, FlaskTrack creates a new organization file for that result.

Destination path

The destination is relative to your organization's FlaskTrack file area.

Example:

analysis/plate-reader

Do not start the path with /.

Avoid path components such as:

..

Filename

Use a filename appropriate for the artifact produced by the upstream block.

Examples:

normalized-results.arrow
summary.json
processed-results.parquet
plate-qc.csv
analysis.png

Output type

The Output block input type must match the upstream artifact.

Examples:

Arrow table → Output configured for Arrow table
File → Output configured for File
JSON → Output configured for JSON

Every run creates its own result

Running the same pipeline again does not rewrite the previous run history.

Each run has its own artifacts and resulting FlaskTrack file records.

This makes it possible to compare results from different executions.

Example:

Pipeline v2
├── Run 101 → normalized-results.arrow
├── Run 102 → normalized-results.arrow
└── Run 103 → normalized-results.arrow

The visible filenames may be the same, but the files belong to different execution records.

Intermediate results versus final outputs

Python and R blocks can generate artifacts while the pipeline runs.

Those artifacts allow downstream blocks to continue processing.

An Output block is what turns a selected artifact into a normal FlaskTrack organization file.

Example:

Source
Python
  ├── intermediate artifact
R
  ├── intermediate artifact
Output
FlaskTrack file

Saving more than one result

Use multiple Output blocks when you want to keep several pipeline results.

Example:

                  ┌──► Output: cleaned-data.parquet
Python:cleaned ───┤
                  └──► R
                    Output: summary.json

Each Output block saves its own configured result.