feat: code is running on hydra 🎉

experiments/train.py

@@ -6,7 +6,6 @@
 from omegaconf import DictConfig
 import torch
 import rootutils
-from hydra.utils import instantiate
 
 rootutils.setup_root(__file__, indicator=".project-root", pythonpath=True)
 
@@ -87,7 +86,7 @@ def train_molbind(config: DictConfig):
     )
 
 
-@hydra.main(config_path="../configs", config_name="train.yaml")
+@hydra.main(version_base="1.3", config_path="../configs", config_name="train.yaml")
 def main(config: DictConfig):
     train_molbind(config)
 

src/molbind/data/dataloaders.py

@@ -2,7 +2,7 @@
 from lightning.pytorch.utilities.combined_loader import CombinedLoader
 from molbind.data.components.tokenizers import SMILES_TOKENIZER, SELFIES_TOKENIZER
 from networkx import Graph
-from typing import Tuple
+from typing import Tuple, Optional
 from torch import Tensor
 
 
@@ -25,9 +25,9 @@ class StringDataset(Dataset):
     def __init__(
         self,
         dataset: Tuple[Tensor, Tensor],
-        central_modality: str,
         modality: str,
-        context_length=256,
+        central_modality: str = "smiles",
+        context_length: Optional[int] = 256,
     ):
         """Dataset for string modalities.
 
@@ -38,13 +38,13 @@ def __init__(
         """
         assert len(dataset) == 2
         assert len(dataset[0]) == len(dataset[1])
-        assert (
-            MODALITY_DATA_TYPES[modality] == str
-        ), "This dataset supports string modalities only."
 
         self.modality = modality
         self.central_modality = central_modality
 
+        assert MODALITY_DATA_TYPES[modality] == str
+        assert MODALITY_DATA_TYPES[central_modality] == str
+
         self.tokenized_central_modality = STRING_TOKENIZERS[central_modality](
             dataset[0],
             padding="max_length",
@@ -62,13 +62,13 @@ def __init__(
         )
 
     def __len__(self):
-        return len(self.tokenized_smiles.input_ids)
+        return len(self.tokenized_central_modality.input_ids)
 
     def __getitem__(self, idx):
         return {
-            "smiles": (
-                self.tokenized_smiles.input_ids[idx],
-                self.tokenized_smiles.attention_mask[idx],
+            self.central_modality: (
+                self.tokenized_central_modality.input_ids[idx],
+                self.tokenized_central_modality.attention_mask[idx],
             ),
             self.modality: (
                 self.tokenized_string.input_ids[idx],
@@ -82,22 +82,22 @@ class GraphDataset(Dataset):
 
 
 def load_combined_loader(
-    central_modality: str,
     data_modalities: dict,
     batch_size: int,
     shuffle: bool,
     num_workers: int,
+    central_modality: str = "smiles",
     drop_last: bool = True,
 ) -> CombinedLoader:
     """Combine multiple dataloaders for different modalities into a single dataloader.
 
     Args:
-        data_modalities (dict): data inputs for each modality as pairs of (SMILES, modality)
+        data_modalities (dict): data inputs for each modality as pairs of (central_modality, modality)
         batch_size (int): batch size for the dataloader
         shuffle (bool): shuffle the dataset
         num_workers (int): number of workers for the dataloader
         drop_last (bool, optional): whether to drop the last batch; defaults to True.
-
+        central_modality (str, optional): central modality to use for the dataset; defaults to "smiles".
     Returns:
         CombinedLoader: a combined dataloader for all the modalities
     """
@@ -106,9 +106,9 @@ def load_combined_loader(
     for modality in [*data_modalities]:
         if MODALITY_DATA_TYPES[modality] == str:
             dataset_instance = StringDataset(
-                dataset=data_modalities,
-                central_modality=central_modality,
+                dataset=data_modalities[modality],
                 modality=modality,
+                central_modality=central_modality,
                 context_length=256,
             )
             loaders[modality] = DataLoader(