Transformer can not be trained on a new dataset!

[CeMarzie]

Traceback (most recent call last): File "/MWPToolkit/run_mwptoolkit.py", line 26, in run_toolkit(args.model, args.dataset, args.task_type, config_dict) File "/MWPToolkit/mwptoolkit/quick_start.py", line 215, in run_toolkit train_with_train_valid_test_split(config) File "/MWPToolkit/mwptoolkit/quick_start.py", line 103, in train_with_train_valid_test_split trainer.fit() File "/MWPToolkit/mwptoolkit/trainer/supervised_trainer.py", line 217, in fit valid_equ_ac, valid_val_ac, valid_total, valid_time_cost = self.evaluate(DatasetType.Valid) File "/MWPToolkit/mwptoolkit/trainer/supervised_trainer.py", line 271, in evaluate batch_val_ac, batch_equ_ac = self._eval_batch(batch) File "/MWPToolkit/mwptoolkit/trainer/supervised_trainer.py", line 148, in _eval_batch test_out, target = self.model.model_test(batch) File "/MWPToolkit/mwptoolkit/model/Seq2Seq/transformer.py", line 154, in model_test _, symbol_outputs, _ = self.forward(src) File "/MWPToolkit/mwptoolkit/model/Seq2Seq/transformer.py", line 100, in forward source_embeddings = self.in_embedder(src).to(self.device) + self.pos_embedder(src).to(self.device) File "/root/anaconda3/envs/toolkit/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl return forward_call(*input, **kwargs) File "/MWPToolkit/mwptoolkit/module/Embedder/basic_embedder.py", line 32, in forward embedding_output = self.embedder(input_seq) File "/root/anaconda3/envs/toolkit/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl return forward_call(*input, **kwargs) File "/root/anaconda3/envs/toolkit/lib/python3.10/site-packages/torch/nn/modules/sparse.py", line 158, in forward return F.embedding( File "/root/anaconda3/envs/toolkit/lib/python3.10/site-packages/torch/nn/functional.py", line 2199, in embedding return torch.embedding(weight, input, padding_idx, scale_grad_by_freq, sparse) RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu! (when checking argument for argument index in method wrapper__index_select)

[CeMarzie]

I appreciate your help in advance.

[LYH-YF]

is it the newest version? @CeMarzie