README.md

metadata

license: mit
tags:
  - dataset
  - math
  - image-text

Math Image Text Data Notes

Dataset summary

A documented Math data-preparation workflow for Image Text records. The bundled rows demonstrate the schema and validation path rather than pretending to be a full training corpus.

Included material

• preprocess.py — loading, cleaning, and split preparation code.
• dataset_infos.json — schema and split metadata.
• metadata_sample.jsonl — small, human-readable records for checking the schema.
• README.md — data card and usage notes.

Processing choices

Stage	Setting
Storage format	csv
Preprocessing	auto ml
Augmentation	autoaugment
Split strategy	temporal
Sampling	balanced
Quality checks	adaptive
Labeling	semi auto

Validation checklist

Before using the prepared data, verify source licenses, duplicates across splits, missing values, label balance, and modality-specific corruption. Record the source version and every filtering rule so a later run can reproduce the same rows.

Intended use

The repository is suitable for testing the data pipeline, adapting it to a documented source, and preparing controlled research splits. Release status: metadata sample; full source data not bundled. The sample is for schema inspection only and should not be reported as a full training corpus.

Risks and limitations

The loader cannot guarantee that an external source is representative, correctly licensed, or free of sensitive information. Users remain responsible for source review, privacy checks, and bias analysis before training or redistribution.

Files

• preprocess.py — primary artifact
• README.md — this documentation
• dataset_infos.json — schema metadata
• metadata_sample.jsonl — schema sample

License

Released under mit. Review the source-data terms separately when this repository is used with external datasets.