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.

Duration

Input Tokens

1199

Output Tokens

206

Cost

$0.00

Context

Input

Label the recipe with dietary and allergen information. Use the following definitions:
- gluten_intolerant: Dish must not contain bread, wheat-based sauces, or anything with gluten
- Vegetarian: Must not contain meat or fish
- Vegan: Must not contain meat, fish, dairy, or eggs
- Halal: Must not contain pork and all meat must be acceptable under halal preparaxtion (assume bacon is not halal)
- Dairy: Contains any dairy products (e.g., cheese, butter, milk, cream)

Expected output

{
  "gluten_intolerant": true,
  "vegetarian": false,
  "vegan": false,
  "halal": true,
  "dairy": true
}

Model output

{
  "gluten_intolerant": true,
  "vegetarian": false,
  "vegan": false,
  "halal": false,
  "dairy": true
}