Source code for

import json
import os
import os.path as osp
from typing import Callable, List, Optional

import numpy as np
import scipy.sparse as sp
import torch

from import Data, InMemoryDataset, download_url

[docs]class Flickr(InMemoryDataset): r"""The Flickr dataset from the `"GraphSAINT: Graph Sampling Based Inductive Learning Method" <>`_ paper, containing descriptions and common properties of images. Args: root (str): Root directory where the dataset should be saved. transform (callable, optional): A function/transform that takes in an :obj:`` object and returns a transformed version. The data object will be transformed before every access. (default: :obj:`None`) pre_transform (callable, optional): A function/transform that takes in an :obj:`` object and returns a transformed version. The data object will be transformed before being saved to disk. (default: :obj:`None`) **STATS:** .. list-table:: :widths: 10 10 10 10 :header-rows: 1 * - #nodes - #edges - #features - #classes * - 89,250 - 899,756 - 500 - 7 """ url = '{}&confirm=t' adj_full_id = '1crmsTbd1-2sEXsGwa2IKnIB7Zd3TmUsy' feats_id = '1join-XdvX3anJU_MLVtick7MgeAQiWIZ' class_map_id = '1uxIkbtg5drHTsKt-PAsZZ4_yJmgFmle9' role_id = '1htXCtuktuCW8TR8KiKfrFDAxUgekQoV7' def __init__(self, root: str, transform: Optional[Callable] = None, pre_transform: Optional[Callable] = None): super().__init__(root, transform, pre_transform), self.slices = torch.load(self.processed_paths[0]) @property def raw_file_names(self) -> List[str]: return ['adj_full.npz', 'feats.npy', 'class_map.json', 'role.json'] @property def processed_file_names(self) -> str: return '' def download(self): path = download_url(self.url.format(self.adj_full_id), self.raw_dir) os.rename(path, osp.join(self.raw_dir, 'adj_full.npz')) path = download_url(self.url.format(self.feats_id), self.raw_dir) os.rename(path, osp.join(self.raw_dir, 'feats.npy')) path = download_url(self.url.format(self.class_map_id), self.raw_dir) os.rename(path, osp.join(self.raw_dir, 'class_map.json')) path = download_url(self.url.format(self.role_id), self.raw_dir) os.rename(path, osp.join(self.raw_dir, 'role.json')) def process(self): f = np.load(osp.join(self.raw_dir, 'adj_full.npz')) adj = sp.csr_matrix((f['data'], f['indices'], f['indptr']), f['shape']) adj = adj.tocoo() row = torch.from_numpy(adj.row).to(torch.long) col = torch.from_numpy(adj.col).to(torch.long) edge_index = torch.stack([row, col], dim=0) x = np.load(osp.join(self.raw_dir, 'feats.npy')) x = torch.from_numpy(x).to(torch.float) ys = [-1] * x.size(0) with open(osp.join(self.raw_dir, 'class_map.json')) as f: class_map = json.load(f) for key, item in class_map.items(): ys[int(key)] = item y = torch.tensor(ys) with open(osp.join(self.raw_dir, 'role.json')) as f: role = json.load(f) train_mask = torch.zeros(x.size(0), dtype=torch.bool) train_mask[torch.tensor(role['tr'])] = True val_mask = torch.zeros(x.size(0), dtype=torch.bool) val_mask[torch.tensor(role['va'])] = True test_mask = torch.zeros(x.size(0), dtype=torch.bool) test_mask[torch.tensor(role['te'])] = True data = Data(x=x, edge_index=edge_index, y=y, train_mask=train_mask, val_mask=val_mask, test_mask=test_mask) data = data if self.pre_transform is None else self.pre_transform(data)[data]), self.processed_paths[0])