Bump tornado from 6.5.5 to 6.5.7

atomgen/models/configuration_atomformer.py

@@ -5,7 +5,7 @@
 from transformers.configuration_utils import PretrainedConfig
 
 
-class AtomformerConfig(PretrainedConfig):
+class AtomformerConfig(PretrainedConfig):  # type: ignore[no-untyped-call]
     r"""
     Configuration of a :class:`~transform:class:`~transformers.AtomformerModel`.
 

atomgen/models/modeling_atomformer.py

@@ -2550,7 +2550,7 @@ def forward(
 class AtomformerPreTrainedModel(PreTrainedModel):  # type: ignore[no-untyped-call]
     """Base class for all transformer models."""
 
-    config_class = AtomformerConfig  # type: ignore[assignment]
+    config_class = AtomformerConfig
     base_model_prefix = "model"
     supports_gradient_checkpointing = True
     _no_split_modules = ["ParallelBlock"]
@@ -2968,9 +2968,9 @@ def __init__(self, config: AtomformerConfig):
 
         if self.problem_type == "regression":
             self.loss_fct = nn.L1Loss()
-        elif self.problem_type == "classification":
+        elif self.problem_type == "classification":  # type: ignore[comparison-overlap]
             self.loss_fct = nn.BCEWithLogitsLoss()
-        elif self.problem_type == "multiclass_classification":
+        elif self.problem_type == "multiclass_classification":  # type: ignore[comparison-overlap]
             self.loss_fct = nn.CrossEntropyLoss()
 
     def forward(
@@ -2989,9 +2989,9 @@ def forward(
 
         loss = None
         if labels is not None:
-            if self.problem_type == "multiclass_classification":
+            if self.problem_type == "multiclass_classification":  # type: ignore[comparison-overlap]
                 labels = labels.long()
-            elif self.problem_type == "classification":
+            elif self.problem_type == "classification":  # type: ignore[comparison-overlap]
                 labels = labels.float()
 
             loss = self.loss_fct(pred.squeeze(), labels.squeeze())

atomgen/models/schnet.py

@@ -10,7 +10,7 @@
 from transformers.modeling_utils import PreTrainedModel
 
 
-class SchNetConfig(PretrainedConfig):
+class SchNetConfig(PretrainedConfig):  # type: ignore[no-untyped-call]
     r"""
     Stores the configuration of a :class:`~transformers.SchNetModel`.
 
@@ -134,7 +134,7 @@ class SchNetPreTrainedModel(PreTrainedModel):  # type: ignore[no-untyped-call]
     simple interface for loading and exporting models.
     """
 
-    config_class = SchNetConfig  # type: ignore[assignment]
+    config_class = SchNetConfig
     base_model_prefix = "model"
     supports_gradient_checkpointing = False
 

atomgen/models/tokengt.py

@@ -2329,7 +2329,7 @@ def forward(
         return out
 
 
-class TransformerConfig(PretrainedConfig):
+class TransformerConfig(PretrainedConfig):  # type: ignore[no-untyped-call]
     """Configuration class to store the configuration of a TokenGT model."""
 
     def __init__(
@@ -2510,7 +2510,7 @@ def custom_forward(*inputs: Any) -> Any:
 class TransformerPreTrainedModel(PreTrainedModel):  # type: ignore[no-untyped-call]
     """Base class for all transformer models."""
 
-    config_class = TransformerConfig  # type: ignore[assignment]
+    config_class = TransformerConfig
     base_model_prefix = "model"
     supports_gradient_checkpointing = True
     _no_split_modules = ["ParallelBlock"]

scripts/training/pretrain_s2ef.py

@@ -178,7 +178,7 @@ def train(args: argparse.Namespace) -> None:
         wandb.login(key=os.environ["WANDB_API_KEY"])
         wandb.init(project=args.project, config=vars(args), name=args.name)
 
-    training_args = TrainingArguments(
+    training_args = TrainingArguments(  # type: ignore[call-arg]
         output_dir=args.output_dir,
         learning_rate=args.learning_rate,
         lr_scheduler_type=args.lr_scheduler_type,