AWeirdDev/all-recipes-sm

--- dataset_info: features: - name: name dtype: string - name: review dtype: string - name: rating dtype: float64 - name: meta struct: - name: active_time dtype: string - name: additional_time dtype: string - name: bake_time dtype: string - name: chill_time dtype: string - name: cook_time dtype: string - name: cool_time dtype: string - name: fry_time dtype: string - name: marinate_time dtype: string - name: prep_time dtype: string - name: rest_time dtype: string - name: servings dtype: string - name: soak_time dtype: string - name: stand_time dtype: string - name: total_time dtype: string - name: yield dtype: string - name: ingredients list: - name: name dtype: string - name: quanity dtype: string - name: unit dtype: string - name: steps sequence: string - name: cooks_note dtype: string - name: editors_note dtype: string - name: nutrition_facts struct: - name: Calories dtype: string - name: Carbs dtype: string - name: Fat dtype: string - name: Protein dtype: string - name: url dtype: string splits: - name: train num_bytes: 3102556 num_examples: 2000 download_size: 1262722 dataset_size: 3102556 configs: - config_name: default data_files: - split: train path: data/train-* tags: - food - recip license: mit task_categories: - text-classification - text-generation - text2text-generation language: - en pretty_name: All Recipes (sm) size_categories: - 1K<n<10K --- # all-recipes-xs (2000) [All Recipes](https://allrecipes.com) dataset (small). ```python from datasets import load_dataset # Load the dataset dataset = load_dataset("AWeirdDev/all-recipes-sm") ``` Alternatively, load with `pickle` from `_frozen.pkl`: ```python import pickle import requests r = requests.get("https://huggingface.co/datasets/AWeirdDev/all-recipes-sm/resolve/main/_frozen.pkl") dataset = pickle.loads(r.content) ``` ## Features **Note:** Empty values are presented as `"unknown"` instead of `None` (normally, unless handled by the 🤗 Datasets library). ```python { "name": "Dutch … Beef", # Name of the recipe "review": "I found this recipe attached…", # Subheading "rating": 5.0, # Overall rating 0 ~ 5 "meta": { # Metadata. May not be present "active_time": None, "additional_time": None, "bake_time": None, "chill_time": None, "cook_time": "4 hrs 5 mins", "cool_time": None, "fry_time": None, "marinate_time": None, "prep_time": "10 mins", "rest_time": None, "servings": "12", "soak_time": None, "stand_time": None, "total_time": "4 hrs 15 mins", "yield": "12 servings" }, "ingredients": [ { "name": "corned beef brisket with spice packet", "quanity": "1", "unit": "(4 pound)" }, ... ], "steps": [ # Steps (unstripped) " Place corned beef brisket…\n", " Combine brown sugar, …\n", " Preheat an outdoor grill …\n", " Place the coated corned …\n" ], "cooks_note": "If there is a…", # Cook's Note (if present) "editors_note": "Nutrition data for this…", # Editor's Note (if present) "nutrition_facts": { "Calories": "251", "Carbs": "16g", "Fat": "13g", "Protein": "13g" }, "url": "https://www.allrecipes.com/recipe/267704/dutch-oven-crunchy-corned-beef/" } ```

原始信息汇总

数据集概述

数据集名称

• 名称: all-recipes-xs (2000)

数据集特征

• 基本特征:
  • name: 字符串类型
  • review: 字符串类型
  • rating: 浮点数类型（64位）
  • meta: 结构化数据，包含多个时间相关的字段，如active_time, cook_time等，均为字符串类型
  • ingredients: 列表类型，包含名称、数量和单位，均为字符串类型
  • steps: 序列类型，字符串
  • cooks_note: 字符串类型
  • editors_note: 字符串类型
  • nutrition_facts: 结构化数据，包含卡路里、碳水化合物、脂肪和蛋白质，均为字符串类型
  • url: 字符串类型

数据集大小

• 训练集:
  • 字节数: 3102556
  • 示例数: 2000
• 下载大小: 1262722
• 数据集大小: 3102556

数据集配置

• 默认配置:
  • 数据文件路径: data/train-*

数据集标签

• 标签:
  • food
  • recip

许可证

• 许可证: MIT

任务类别

• 任务类别:
  • text-classification
  • text-generation
  • text2text-generation

语言

• 语言: en

数据集别名

• 别名: All Recipes (sm)

大小类别

• 大小类别: 1K<n<10K