Data Normalization

Data processing and normalization tasks evaluate structured output from messy prose and different structures. This capability is essential for catalogue and product pipelines where data needs to be extracted from unstructured text and formatted consistently. Models must understand the desired output format and extract relevant information accurately.

50
Input Tokens
0
Output Tokens
0
Duration
0
Cost
0
Context
Input
From the extracted description. Extra all info about the clothing
Expected output
{
  "top_piece": {
    "type": "shirt",
    "color": "blue",
    "size": "M"
  },
  "outerwear": {
    "type": "blazer",
    "color": "navy",
    "size": "L"
  },
  "bottom_piece": {
    "type": "chinos",
    "color": "beige",
    "size": "M"
  },
  "shoes": {
    "type": "brogues",
    "color": "brown",
    "size": "M"
  },
  "watch": {
    "material": "stainless steel",
    "strap_color": "black"
  },
  "hat": {
    "type": "flat cap",
    "color": "gray",
    "material": "wool"
  }
}
Model output
{
  "top_piece": {
    "type": "linen shirt",
    "color": "medium blue",
    "size": "M"
  },
  "outerwear": {
    "type": "wool blazer",
    "color": "navy",
    "size": "L"
  },
  "bottom_piece": {
    "type": "chinos",
    "color": "medium beige",
    "size": "M"
  },
  "shoes": {
    "type": "leather brogues",
    "color": "brown",
    "size": "M"
  },
  "watch": {
    "material": "stainless steel",
    "strap_color": "black"
  },
  "hat": {
    "type": "flat cap",
    "color": "gray",
    "material": "wool"
  }
}