Examples

This page collects the public sample files shipped with the documentation.

Quickstart Example

The smallest runnable example is:

It uses:

  • format args

  • two stages

  • simple filename derivation through aliases

args Example

The more representative args-based sample is:

It demonstrates:

  • pipelines and tasks

  • resource limits with limit cpu, ...

  • scalar aliases such as {sample} and {tmp}

varlists Example

The varlists sample set is:

These files demonstrate:

  • format varlists

  • [list] with {#2}, ...

  • list-typed alias generation

  • run_list fan-out

  • {[list_alias]} join-back in a later stage

Log Management Example

These files demonstrate:

  • [alias] for filename construction

  • parallel upload on [stage.3]

Media Processing Example

These files demonstrate:

  • one stage that runs two ffmpeg commands in parallel

  • one transcription stage

  • one cleanup stage

Machine Learning Examples

Batch Inference from CSV

The batch inference sample set is:

This example fits workloads where each row describes one sample, one model, and one prompt or config file.

Feature Extraction with varlists and run_list

The feature-pack sample set is:

These files demonstrate:

  • format varlists

  • [list] with {#2}, ...

  • run_list fan-out for one file per image

  • {[feature_files]} join-back for later packing

  • a cleanup stage that removes intermediates

This is the recommended sample to read if you want a machine-learning-flavored example of varlists plus run_list.

RNA-seq Paired-End Example

A bioinformatics domain-specific sample is:

This example models one paired-end RNA-seq library per pipeline.

It assumes that each input row is a read-1 filename such as:

LIB001_1.fastq.gz

Recommended invocation:

minigate -f docs/source/minigate/examples/rnaseq.pipeline *_1.fastq.gz

  • if you pass *.fastq.gz, both read-1 and read-2 files become independent input rows

  • *_1.fastq.gz is the intended shell pattern for this example