The matrix dimension is very large, resulting in an attempt to allocate a very large array resulting in low memory.
It is recommended to process data in chunks: Developers can consider dividing data into smaller chunks for processing. This avoids allocating a large amount of memory at once.
deeph-inference --config inference.ini
User config name: ['inference.ini']
~~~~~~~ 2.get_local_coordinate
~~~~~~~ 3.get_pred_Hamiltonian
~~~~~~~ 4.rotate_back
~~~~~~~ 5.sparse_calc, command:
xxx/sparse_calc.jl --input_dir xxx/get_S_process --config
####### Begin 1.parse_Overlap
Output subdirectories: OUT.ABACUS
Traceback (most recent call last):
File "xxx/deeph-inference", line 8, in <module>
sys.exit(main())
File "xxx/deeph/scripts/inference.py", line 97, in main
abacus_parse(OLP_dir, work_dir, data_name=f'OUT.{abacus_suffix}', only_S=True)
File "xxx/deeph/preprocess/abacus_get_data.py", line 247, in abacus_parse
overlap_dict, tmp = parse_matrix(os.path.join(input_path, "SR.csr"), 1)
File "xxx/deeph/preprocess/abacus_get_data.py", line 215, in parse_matrix
hamiltonian_cur = csr_matrix((np.array(line2).astype(float), np.array(line3).astype(int),
File "xxx/site-packages/scipy/sparse/_compressed.py", line 1051, in toarray
out = self._process_toarray_args(order, out)
File "xxx/site-packages/scipy/sparse/_base.py", line 1298, in _process_toarray_args
return np.zeros(self.shape, dtype=self.dtype, order=order)
numpy.core._exceptions._ArrayMemoryError: Unable to allocate 6.03 TiB for an array with shape (910224, 910224) and data type float64
The matrix dimension is very large, resulting in an attempt to allocate a very large array resulting in low memory. It is recommended to process data in chunks: Developers can consider dividing data into smaller chunks for processing. This avoids allocating a large amount of memory at once.
deeph-inference --config inference.ini
User config name: ['inference.ini']