add recipe-100k and recipe-200k datasets

Author: yifanfeng97

dhg/data/__init__.py

@@ -19,6 +19,7 @@
 from .citeseer import Citeseer, CiteseerBiGraph
 from .pubmed import Pubmed, PubmedBiGraph
 from .imdb import IMDB4k
+from .recipe import Recipe100k, Recipe200k
 
 __all__ = [
     "BaseData",
@@ -50,4 +51,6 @@
     "HouseCommittees",
     "News20",
     "IMDB4k",
+    "Recipe100k",
+    "Recipe200k"
 ]

dhg/data/news.py

@@ -8,7 +8,7 @@
 
 class News20(BaseData):
     r"""The 20 Newsgroups dataset is a newspaper network dataset for vertex classification task. 
-    The node features are the TF-IDF representations of news messages.
+    The vertex features are the TF-IDF representations of news messages.
     More details see the `YOU ARE ALLSET: A MULTISET LEARNING FRAMEWORK FOR HYPERGRAPH NEURAL NETWORKS <https://openreview.net/pdf?id=hpBTIv2uy_E>`_ paper.
     
     The content of the 20 Newsgroups dataset includes the following:

dhg/data/recipe.py

@@ -0,0 +1,95 @@
+from typing import Optional
+from functools import partial
+
+from dhg.datapipe import load_from_pickle, norm_ft, to_tensor, to_long_tensor, to_bool_tensor
+
+from .base import BaseData
+
+
+class Recipe100k(BaseData):
+    r"""The Recipe100k dataset is a recipe-ingredient network dataset for vertex classification task. 
+    The vertex features are the bag of words from the sentence that making the recipe. 
+    Hyperedges are the ingredients of the recipe or the Keywords for food preparation steps. 
+    The original dataset is created in `SHARE: a System for Hierarchical Assistive Recipe Editing <https://arxiv.org/pdf/2105.08185.pdf>`_ paper.
+    
+    The content of the Recipe100k dataset includes the following:
+
+    - ``num_classes``: The number of classes: :math:`8`.
+    - ``num_vertices``: The number of vertices: :math:`101,585`.
+    - ``num_edges``: The number of edges: :math:`12,387`.
+    - ``dim_features``: The dimension of features: :math:`2,254`.
+    - ``features``: The vertex feature matrix. ``torch.Tensor`` with size :math:`(101,585 \times 2,254)`.
+    - ``edge_list``: The edge list. ``List`` with length :math:`12,387`.
+    - ``labels``: The label list. ``torch.LongTensor`` with size :math:`(101,585, )`.
+
+    Args:
+        ``data_root`` (``str``, optional): The ``data_root`` has stored the data. If set to ``None``, this function will auto-download from server and save into the default direction ``~/.dhg/datasets/``. Defaults to ``None``.
+    """
+
+    def __init__(self, data_root: Optional[str] = None) -> None:
+        super().__init__("recipe-100k-v2", data_root)
+        self._content = {
+            "num_classes": 8,
+            "num_vertices": 101585,
+            "num_edges": 12387,
+            "dim_features": 2254,
+            "features": {
+                "upon": [{"filename": "features.pkl", "md5": "4fdd76cd4108fd07bdd62368067c1eaf"}],
+                "loader": load_from_pickle,
+                "preprocess": [to_tensor,],
+            },
+            "edge_list": {
+                "upon": [{"filename": "edge_list.pkl", "md5": "3dc1d8fe7a0f91b5c56057500bda9021"}],
+                "loader": load_from_pickle,
+            },
+            "labels": {
+                "upon": [{"filename": "labels.pkl", "md5": "bd8a3bcaef27a58c6d1d5def255c5065"}],
+                "loader": load_from_pickle,
+                "preprocess": [to_long_tensor],
+            },
+        }
+
+
+class Recipe200k(BaseData):
+    r"""The Recipe200k dataset is a recipe-ingredient network dataset for vertex classification task. 
+    The vertex features are the bag of words from the sentence that making the recipe. 
+    Hyperedges are the ingredients of the recipe or the Keywords for food preparation steps. 
+    The original dataset is created in `SHARE: a System for Hierarchical Assistive Recipe Editing <https://arxiv.org/pdf/2105.08185.pdf>`_ paper.
+    
+    The content of the Recipe200k dataset includes the following:
+
+    - ``num_classes``: The number of classes: :math:`8`.
+    - ``num_vertices``: The number of vertices: :math:`240,094`.
+    - ``num_edges``: The number of edges: :math:`18,129`.
+    - ``dim_features``: The dimension of features: :math:`3,200`.
+    - ``features``: The vertex feature matrix. ``torch.Tensor`` with size :math:`(240,094 \times 3,200)`.
+    - ``edge_list``: The edge list. ``List`` with length :math:`18,129`.
+    - ``labels``: The label list. ``torch.LongTensor`` with size :math:`(240,094, )`.
+
+    Args:
+        ``data_root`` (``str``, optional): The ``data_root`` has stored the data. If set to ``None``, this function will auto-download from server and save into the default direction ``~/.dhg/datasets/``. Defaults to ``None``.
+    """
+
+    def __init__(self, data_root: Optional[str] = None) -> None:
+        super().__init__("recipe-200k-v2", data_root)
+        self._content = {
+            "num_classes": 8,
+            "num_vertices": 240094,
+            "num_edges": 18129,
+            "dim_features": 3200,
+            "features": {
+                "upon": [{"filename": "features.pkl", "md5": "a5df55a3e9591d7389f6ea5f09a483f4"}],
+                "loader": load_from_pickle,
+                "preprocess": [to_tensor,],
+            },
+            "edge_list": {
+                "upon": [{"filename": "edge_list.pkl", "md5": "163ad784e35e56650fc22658d3e88767"}],
+                "loader": load_from_pickle,
+            },
+            "labels": {
+                "upon": [{"filename": "labels.pkl", "md5": "05bee03f1c5383f0cde5ea879be090af"}],
+                "loader": load_from_pickle,
+                "preprocess": [to_long_tensor],
+            },
+        }
+

docs/source/api/data.rst

@@ -64,6 +64,8 @@ Hypergraph Datasets
     dhg.data.DBLP4k
     dhg.data.DBLP8k
     dhg.data.IMDB4k
+    dhg.data.Recipe100k
+    dhg.data.Recipe200k
 
 
 **Welcome to contribute datasets!**