README.md

metadata

license: mit
task_categories:
  - image-to-text
  - text-generation
language:
  - en
tags:
  - recipes
  - food
  - multimodal
  - rag
size_categories:
  - 10K<n<100K

Recipe PDF Dataset

A multimodal dataset of 10K+ recipes rendered as PDF images with full metadata.

Dataset Description

Each sample contains:

• image: Recipe rendered as a styled PDF page (PNG, ~1654x2339px)
• name: Recipe title
• description: Recipe description
• ingredients: List of ingredients
• steps: Cooking instructions
• nutrition: Nutritional values (calories, fat%, sugar%, sodium%, protein%, sat.fat%, carbs%)
• random_reviews: User reviews
• minutes: Cooking time
• tags: Recipe categories
• n_steps / n_ingredients: Counts

Usage

from datasets import load_dataset

dataset = load_dataset("your-username/recipe-pdf-dataset")
sample = dataset["train"][0]

# Access image
sample["image"].show()

# Access metadata
print(sample["name"])
print(sample["ingredients"])

Source

Generated from Food.com Recipes using custom PDF rendering with varied layouts, fonts, and color palettes.

Stats

Split	Samples
train	10,096