search

qitianwu / DIFFormer

The official implementation for ICLR23 spotlight paper "DIFFormer: Scalable (Graph) Transformers Induced by Energy Constrained Diffusion"
302 stars 32 forks source link

Cannot load pokec dataset #7

Closed Xi-yuanWang closed 1 year ago

Xi-yuanWang commented 1 year ago

Hello. It seems that scipy cannot read the mat file of the pokec dataset. Could you please help me?

Traceback (most recent call last): File "/home/wangxiyuan/DIFFormer/node classification/main-batch.py", line 43, in dataset = load_dataset(args.data_dir, args.dataset, args.sub_dataset) File "/home/wangxiyuan/DIFFormer/node classification/dataset.py", line 109, in load_dataset dataset = load_pokec_mat(data_dir) File "/home/wangxiyuan/DIFFormer/node classification/dataset.py", line 312, in load_pokec_mat fulldata = scipy.io.loadmat(f'{data_dir}pokec.mat') File "/home/wangxiyuan/miniconda3/lib/python3.10/site-packages/scipy/io/matlab/mio.py", line 225, in loadmat MR, = mat_reader_factory(f, **kwargs) File "/home/wangxiyuan/miniconda3/lib/python3.10/site-packages/scipy/io/matlab/_mio.py", line 74, in mat_reader_factory mjv, mnv = _get_matfile_version(byte_stream) File "/home/wangxiyuan/miniconda3/lib/python3.10/site-packages/scipy/io/matlab/_miobase.py", line 251, in _get_matfile_version raise ValueError('Unknown mat file type, version %s, %s' % ret) ValueError: Unknown mat file type, version 32, 99

qitianwu commented 1 year ago

Hi, have you followed the requirements.txt for installing the correct version of the scipy package

Xi-yuanWang commented 1 year ago

Sorry for my late reply. Yes. I use scipy 1.6.2. Could you please provide npy format data?

qitianwu commented 1 year ago

Sorry for the late reply, the original data of pokec is provided in the mat format. So I doubt the issue is caused by the mismatch of packages. Our dataloader follows the implementation

https://github.com/CUAI/Non-Homophily-Benchmarks