runtime error

Exit code: 1. Reason: instantiate_classes parent[key] = component.instantiate_classes(value) ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^ File "/usr/local/lib/python3.13/site-packages/jsonargparse/_typehints.py", line 611, in instantiate_classes value[num] = adapt_typehints( ~~~~~~~~~~~~~~~^ val, ^^^^ ...<4 lines>... logger=self.logger, ^^^^^^^^^^^^^^^^^^^ ) ^ File "/usr/local/lib/python3.13/site-packages/jsonargparse/_typehints.py", line 826, in adapt_typehints raise_union_unexpected_value(typehint, val, vals) ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.13/site-packages/jsonargparse/_typehints.py", line 710, in raise_union_unexpected_value raise ValueError( ...<2 lines>... ) from exceptions[0] ValueError: Does not validate against any of the Union subtypes Subtypes: (<class 'torchgeo.trainers.base.BaseTask'>, <class 'NoneType'>) Errors: - UnetDecoder.__init__() got an unexpected keyword argument 'use_batchnorm' - Expected a <class 'NoneType'> Given value type: <class 'jsonargparse._namespace.Namespace'> Given value: Namespace(class_path='terratorch.tasks.SemanticSegmentationTask', init_args=Namespace(model_args={'backbone': 'prithvi_eo_v2_300', 'backbone_pretrained': True, 'backbone_bands': ['BLUE', 'GREEN', 'RED', 'NIR_NARROW', 'SWIR_1', 'SWIR_2'], 'necks': [{'name': 'SelectIndices', 'indices': [5, 11, 17, 23]}, {'name': 'ReshapeTokensToImage'}, {'name': 'LearnedInterpolateToPyramidal'}], 'decoder': 'UNetDecoder', 'decoder_channels': [512, 256, 128, 64], 'num_classes': 2}, model_factory='EncoderDecoderFactory', model=None, loss='ce', aux_heads=None, aux_loss=None, class_weights=None, ignore_index=-1, lr=0.001, optimizer=None, optimizer_hparams=None, scheduler=None, scheduler_hparams=None, freeze_backbone=False, freeze_decoder=False, freeze_head=False, plot_on_val=False, class_names=['Not burned', 'Burn scar'], tiled_inference_parameters=None, test_dataloaders_names=None, lr_overrides=None))

Container logs: