Having Issues re training the model with the existing data provided #6
Description
I am getting the following error on re training the uspto data that has been provided in the link: https://drive.google.com/drive/folders/1lZOLRGyZy18EVow7gyxtKWvs_yuwlIE3?usp=sharing
The atoms count error shouldn't appear at all since the data is the existing one that the model is already trained on. Any thought?
[rank0]: Traceback (most recent call last):
[rank0]: File "/usr/local/bin/unicore-train", line 8, in <module>
[rank0]: sys.exit(cli_main())
[rank0]: File "/usr/local/lib/python3.10/dist-packages/unicore_cli/train.py", line 418, in cli_main
[rank0]: distributed_utils.call_main(args, main)
[rank0]: File "/usr/local/lib/python3.10/dist-packages/unicore/distributed/utils.py", line 186, in call_main
[rank0]: distributed_main(int(os.environ["LOCAL_RANK"]), main, args, kwargs)
[rank0]: File "/usr/local/lib/python3.10/dist-packages/unicore/distributed/utils.py", line 160, in distributed_main
[rank0]: main(args, **kwargs)
[rank0]: File "/usr/local/lib/python3.10/dist-packages/unicore_cli/train.py", line 105, in main
[rank0]: extra_state, epoch_itr = checkpoint_utils.load_checkpoint(
[rank0]: File "/usr/local/lib/python3.10/dist-packages/unicore/checkpoint_utils.py", line 223, in load_checkpoint
[rank0]: extra_state, epoch_itr = trainer.load_checkpoint(
[rank0]: File "/usr/local/lib/python3.10/dist-packages/unicore/trainer.py", line 433, in load_checkpoint
[rank0]: epoch_itr = self.get_train_iterator(
[rank0]: File "/usr/local/lib/python3.10/dist-packages/unicore/trainer.py", line 508, in get_train_iterator
[rank0]: self.reset_dummy_batch(batch_iterator.first_batch)
[rank0]: File "/usr/local/lib/python3.10/dist-packages/unicore/data/iterators.py", line 243, in first_batch
[rank0]: return self.collate_fn([self.dataset[i] for i in self.frozen_batches[0]])
[rank0]: File "/usr/local/lib/python3.10/dist-packages/unicore/data/iterators.py", line 243, in <listcomp>
[rank0]: return self.collate_fn([self.dataset[i] for i in self.frozen_batches[0]])
[rank0]: File "/usr/local/lib/python3.10/dist-packages/unicore/data/base_wrapper_dataset.py", line 18, in __getitem__
[rank0]: return self.dataset[index]
[rank0]: File "/usr/local/lib/python3.10/dist-packages/unicore/data/nested_dictionary_dataset.py", line 69, in __getitem__
[rank0]: return OrderedDict((k, ds[index]) for k, ds in self.defn.items())
[rank0]: File "/usr/local/lib/python3.10/dist-packages/unicore/data/nested_dictionary_dataset.py", line 69, in <genexpr>
[rank0]: return OrderedDict((k, ds[index]) for k, ds in self.defn.items())
[rank0]: File "/usr/local/lib/python3.10/dist-packages/unimol/data/molecule_dataset.py", line 116, in __getitem__
[rank0]: return self.__getitem_cached__(self.epoch, idx)
[rank0]: File "/usr/local/lib/python3.10/dist-packages/unimol/data/molecule_dataset.py", line 121, in __getitem_cached__
[rank0]: data = self.dataset[idx]
[rank0]: File "/usr/local/lib/python3.10/dist-packages/unimol/data/key_dataset.py", line 27, in __getitem__
[rank0]: return self.__cached_item__(idx, self.epoch)
[rank0]: File "/usr/local/lib/python3.10/dist-packages/unimol/data/key_dataset.py", line 24, in __cached_item__
[rank0]: return self.dataset[idx][self.key]
[rank0]: File "/content/NAG2G/NAG2G/data/graphormer_dataset.py", line 132, in __getitem__
[rank0]: return self.__getitem_cached__(self.epoch, index)
[rank0]: File "/content/NAG2G/NAG2G/data/graphormer_dataset.py", line 137, in __getitem_cached__
[rank0]: reactant = self.reactant_dataset[index]
[rank0]: File "/content/NAG2G/NAG2G/data/graphormer_dataset.py", line 67, in __getitem__
[rank0]: return self.__getitem_cached__(self.epoch, index)
[rank0]: File "/content/NAG2G/NAG2G/data/graphormer_dataset.py", line 71, in __getitem_cached__
[rank0]: smiles = self.dataset[index]
[rank0]: File "/content/NAG2G/NAG2G/data/random_smiles_dataset.py", line 75, in __getitem__
[rank0]: return self.__getitem_cached__(self.epoch, index)
[rank0]: File "/content/NAG2G/NAG2G/data/random_smiles_dataset.py", line 85, in __getitem_cached__
[rank0]: nm = Chem.RenumberAtoms(reactant_mol, list_reactant)
[rank0]: ValueError: atomCounts shorter than the number of atoms