SimpleDet Docs

Datasets

SimpleDet can inspect several dataset formats, but the suite plus pipeline workflow is still centered on COCO-style annotations.

Overview Quickstart Core Concepts API Reference

Use this page to decide which dataset formats are suitable for exploration and which ones must be converted before full training.

Supported loader formats

Format	Status	Notes
`coco`	Implemented	Best aligned with the training pipeline
`csv`	Implemented	Flat annotation tables for exploration
`json` / `jsonl` / `ndjson`	Implemented	Line-delimited records
`yolo`	Implemented	Use for loading, not for the pipeline runtime
`voc`	Implemented	Pascal VOC XML support

Recommended layout

dataset_root/
  annotations/
    train.json
    val.json
    test.json
  imgs/
    image_0001.tif
    image_0002.tif

COCO example

{
  "images": [{"id": 1, "file_name": "image_0001.png", "width": 1024, "height": 1024}],
  "annotations": [{"id": 1, "image_id": 1, "category_id": 0, "bbox": [100, 120, 50, 80], "area": 4000, "iscrowd": 0}],
  "categories": [{"id": 0, "name": "vessel"}]
}

Practical rule

Use generic loaders for data exploration and the lightweight helpers. Convert to COCO JSON before using the native runtime helpers with detector_spec=....