Traceback (most recent call last):
File "D:\PycharmProjects\AlphaRouter\debug_pretrain.py", line 99, in
_work(**params)
File "D:\PycharmProjects\AlphaRouter\debug_pretrain.py", line 33, in _work
run_pretrain(args)
File "D:\PycharmProjects\AlphaRouter\src\run.py", line 92, in run_pretrain
File "D:\PycharmProjects\AlphaRouter\src\pretrain_model\pretrainer_module.py", line 104, in run
File "D:\PycharmProjects\AlphaRouter\src\pretrain_model\pretrainer_module.py", line 183, in _train_one_epoch
File "D:\PycharmProjects\AlphaRouter\src\env\np_vec\tsp_np_vec_env.py", line 46, in reset
self.xy = self._make_problems(self.num_env, self.num_nodes)
File "D:\PycharmProjects\AlphaRouter\src\env\np_vec\tsp_np_vec_env.py", line 29, in _make_problems
xy = make_cord(num_rollouts, 0, num_nodes)
File "D:\PycharmProjects\AlphaRouter\src\common\data_manipulator.py", line 8, in make_cord
node_cord = np.random.rand(num_rollouts, num_nodes, 2).astype(np.float32)
File "mtrand.pyx", line 1202, in numpy.random.mtrand.RandomState.rand
File "mtrand.pyx", line 436, in numpy.random.mtrand.RandomState.random_sample
File "_common.pyx", line 307, in numpy.random._common.double_fill
numpy.core._exceptions._ArrayMemoryError: Unable to allocate 1.56 MiB for an array with shape (1024, 100, 2) and data type float64
It seems to occur as numpy uses f.64 as default type.
Traceback (most recent call last): File "D:\PycharmProjects\AlphaRouter\debug_pretrain.py", line 99, in
_work(**params)
File "D:\PycharmProjects\AlphaRouter\debug_pretrain.py", line 33, in _work
run_pretrain(args)
File "D:\PycharmProjects\AlphaRouter\src\run.py", line 92, in run_pretrain
File "D:\PycharmProjects\AlphaRouter\src\pretrain_model\pretrainer_module.py", line 104, in run
File "D:\PycharmProjects\AlphaRouter\src\pretrain_model\pretrainer_module.py", line 183, in _train_one_epoch
File "D:\PycharmProjects\AlphaRouter\src\env\np_vec\tsp_np_vec_env.py", line 46, in reset self.xy = self._make_problems(self.num_env, self.num_nodes) File "D:\PycharmProjects\AlphaRouter\src\env\np_vec\tsp_np_vec_env.py", line 29, in _make_problems xy = make_cord(num_rollouts, 0, num_nodes) File "D:\PycharmProjects\AlphaRouter\src\common\data_manipulator.py", line 8, in make_cord node_cord = np.random.rand(num_rollouts, num_nodes, 2).astype(np.float32) File "mtrand.pyx", line 1202, in numpy.random.mtrand.RandomState.rand File "mtrand.pyx", line 436, in numpy.random.mtrand.RandomState.random_sample File "_common.pyx", line 307, in numpy.random._common.double_fill numpy.core._exceptions._ArrayMemoryError: Unable to allocate 1.56 MiB for an array with shape (1024, 100, 2) and data type float64
It seems to occur as numpy uses f.64 as default type.