add dummies and we should be good to go!z

Author: ArthurZucker

src/transformers/utils/dummy_pt_objects.py

@@ -4981,6 +4981,41 @@ def __init__(self, *args, **kwargs):
         requires_backends(self, ["torch"])
 
 
+class HeliumForCausalLM(metaclass=DummyObject):
+    _backends = ["torch"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+
+class HeliumForSequenceClassification(metaclass=DummyObject):
+    _backends = ["torch"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+
+class HeliumForTokenClassification(metaclass=DummyObject):
+    _backends = ["torch"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+
+class HeliumModel(metaclass=DummyObject):
+    _backends = ["torch"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+
+class HeliumPreTrainedModel(metaclass=DummyObject):
+    _backends = ["torch"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+
 class HieraBackbone(metaclass=DummyObject):
     _backends = ["torch"]