理解 Potato 数据格式

Potato 使用 JSON 和 JSONL 格式处理输入数据和输出标注。本指南涵盖所有数据类型的格式规范、示例和最佳实践。

输入数据格式

JSON Lines (JSONL) - 推荐

每行一个 JSON 对象：

json

{"id": "001", "text": "First document text here."}
{"id": "002", "text": "Second document text here."}
{"id": "003", "text": "Third document text here."}

优势：

流式处理（内存高效）
便于追加数据
单行损坏不会影响整个文件

JSON 数组

标准 JSON 数组：

json

[
  {"id": "001", "text": "First document."},
  {"id": "002", "text": "Second document."},
  {"id": "003", "text": "Third document."}
]

配置：

yaml

data_files:
  - data/items.json

文本标注数据

基本文本

json

{"id": "doc_001", "text": "The product quality exceeded my expectations."}

带元数据

json

{
  "id": "review_001",
  "text": "Great product, fast shipping!",
  "metadata": {
    "source": "amazon",
    "date": "2024-01-15",
    "author": "user123",
    "rating": 5
  }
}

带预标注

json

{
  "id": "ner_001",
  "text": "Apple announced new products in Cupertino.",
  "pre_annotations": {
    "entities": [
      {"start": 0, "end": 5, "label": "ORG", "text": "Apple"},
      {"start": 31, "end": 40, "label": "LOC", "text": "Cupertino"}
    ]
  }
}

配置：

yaml

data_files:
  - data/texts.json
 
item_properties:
  id_key: id
  text_key: text

图像标注数据

本地图像

json

{
  "id": "img_001",
  "image_path": "/data/images/photo_001.jpg",
  "caption": "Street scene in Paris"
}

远程图像

json

{
  "id": "img_002",
  "image_url": "https://example.com/images/photo.jpg"
}

带边界框

json

{
  "id": "detection_001",
  "image_path": "/images/street.jpg",
  "pre_annotations": {
    "objects": [
      {"bbox": [100, 150, 200, 300], "label": "person"},
      {"bbox": [350, 200, 450, 280], "label": "car"}
    ]
  }
}

配置：

yaml

data_files:
  - data/images.json
 
item_properties:
  id_key: id
  image_key: image_path  # or image_url

音频标注数据

本地音频

json

{
  "id": "audio_001",
  "audio_path": "/data/audio/recording.wav",
  "duration": 45.5,
  "transcript": "Hello, how are you today?"
}

带分段

json

{
  "id": "audio_002",
  "audio_path": "/audio/meeting.mp3",
  "segments": [
    {"start": 0.0, "end": 5.5, "speaker": "Speaker1"},
    {"start": 5.5, "end": 12.0, "speaker": "Speaker2"}
  ]
}

配置：

yaml

data_files:
  - data/audio.json
 
item_properties:
  audio_key: audio_path
  text_key: transcript

多模态数据

文本 + 图像

json

{
  "id": "mm_001",
  "text": "What is shown in this image?",
  "image_path": "/images/scene.jpg"
}

文本 + 音频

json

{
  "id": "mm_002",
  "text": "Transcribe this audio:",
  "audio_path": "/audio/clip.wav",
  "reference_transcript": "Expected transcription here"
}

输出标注格式

基本输出

json

{
  "id": "doc_001",
  "text": "Great product!",
  "annotations": {
    "sentiment": "Positive",
    "confidence": 5
  },
  "annotator": "user123",
  "timestamp": "2024-11-05T10:30:00Z"
}

Span 标注

json

{
  "id": "ner_001",
  "text": "Apple CEO Tim Cook visited Paris.",
  "annotations": {
    "entities": [
      {"start": 0, "end": 5, "label": "ORG", "text": "Apple"},
      {"start": 10, "end": 18, "label": "PERSON", "text": "Tim Cook"},
      {"start": 27, "end": 32, "label": "LOC", "text": "Paris"}
    ]
  }
}

多标注者

json

{
  "id": "item_001",
  "text": "Sample text",
  "annotations": [
    {
      "annotator": "ann1",
      "labels": {"sentiment": "Positive"},
      "timestamp": "2024-11-05T10:00:00Z"
    },
    {
      "annotator": "ann2",
      "labels": {"sentiment": "Positive"},
      "timestamp": "2024-11-05T11:00:00Z"
    }
  ],
  "aggregated": {
    "sentiment": "Positive",
    "agreement": 1.0
  }
}

配置参考

yaml

data_files:
  - data/items.json
 
item_properties:
  id_key: id
  text_key: text
  image_key: image_path
  audio_key: audio_path

最佳实践

始终包含 ID：用于追踪的唯一标识符
大数据集使用 JSONL：更好的内存效率
加载前验证：检查 JSON 语法
包含元数据：来源、日期、作者有助于调试
一致的字段名：方便下游处理

完整的数据格式文档请参阅 /docs/core-concepts/data-formats。