ValueError: dictionary update sequence element #0 has length 1; 2 is required

[OttoTim]

Hey,

I followed your steps and executed

python finetune.py  --accelerator 'gpu' --devices 1  --num_nodes 1 --strategy 'ddp' \
           --max_epochs 50 \
           --task_type ${TASK_TYPE} \
           --default_root_dir ${LOG_FOLDER} \
           --data_dir ${PROCESSED_DATA_DIR}  \
           --dataset ${SPLIT_FILE} \
           --train_batch_size 32  \
           --val_batch_size 32  \
           --model_name_or_path ${PRETRAIN_MODEL_PATH} \
           --num_labels ${NUM_LABELS} \
           --random_seed 0

but I encountered the following error:

[rank0]: Traceback (most recent call last):
[rank0]: File "/data/otto/table-union-search-benchmark/approaches/tabsketchfm/finetune.py", line 188, in
[rank0]: cli_main()
[rank0]: ~~~~~~~~^^
[rank0]: File "/data/otto/table-union-search-benchmark/approaches/tabsketchfm/finetune.py", line 174, in cli_main
[rank0]: trainer_ddp.fit(model, data_module)
[rank0]: ~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^
[rank0]: File "/home/otto/.conda/envs/latest/lib/python3.13/site-packages/pytorch_lightning/trainer/trainer.py", line 603, in fit
[rank0]: call._call_and_handle_interrupt(
[rank0]: ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^
[rank0]: self, self._fit_impl, model, train_dataloaders, val_dataloaders, datamodule, ckpt_path
[rank0]: ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[rank0]: )
[rank0]: ^
[rank0]: File "/home/otto/.conda/envs/latest/lib/python3.13/site-packages/pytorch_lightning/trainer/call.py", line 36, in _call_and_handle_interrupt
[rank0]: return trainer.strategy.launcher.launch(trainer_fn, *args, trainer=trainer, **kwargs)
[rank0]: ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[rank0]: File "/home/otto/.conda/envs/latest/lib/python3.13/site-packages/pytorch_lightning/strategies/launchers/subprocess_script.py", line 90, in launch
[rank0]: return function(*args, **kwargs)
[rank0]: File "/home/otto/.conda/envs/latest/lib/python3.13/site-packages/pytorch_lightning/trainer/trainer.py", line 645, in _fit_impl
[rank0]: self._run(model, ckpt_path=self.ckpt_path)
[rank0]: ~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[rank0]: File "/home/otto/.conda/envs/latest/lib/python3.13/site-packages/pytorch_lightning/trainer/trainer.py", line 1086, in _run
[rank0]: self._log_hyperparams()
[rank0]: ~~~~~~~~~~~~~~~~~~~~~^^
[rank0]: File "/home/otto/.conda/envs/latest/lib/python3.13/site-packages/pytorch_lightning/trainer/trainer.py", line 1156, in _log_hyperparams
[rank0]: logger.save()
[rank0]: ~~~~~~~~~~~^^
[rank0]: File "/home/otto/.conda/envs/latest/lib/python3.13/site-packages/lightning_utilities/core/rank_zero.py", line 42, in wrapped_fn
[rank0]: return fn(*args, **kwargs)
[rank0]: File "/home/otto/.conda/envs/latest/lib/python3.13/site-packages/pytorch_lightning/loggers/tensorboard.py", line 287, in save
[rank0]: save_hparams_to_yaml(hparams_file, self.hparams)
[rank0]: ~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[rank0]: File "/home/otto/.conda/envs/latest/lib/python3.13/site-packages/pytorch_lightning/core/saving.py", line 407, in save_hparams_to_yaml
[rank0]: yaml.dump(v)
[rank0]: ~~~~~~~~~^^^
[rank0]: File "/home/otto/.conda/envs/latest/lib/python3.13/site-packages/yaml/init.py", line 253, in dump
[rank0]: return dump_all([data], stream, Dumper=Dumper, **kwds)
[rank0]: File "/home/otto/.conda/envs/latest/lib/python3.13/site-packages/yaml/init.py", line 241, in dump_all
[rank0]: dumper.represent(data)
[rank0]: ~~~~~~~~~~~~~~~~^^^^^^
[rank0]: File "/home/otto/.conda/envs/latest/lib/python3.13/site-packages/yaml/representer.py", line 27, in represent
[rank0]: node = self.represent_data(data)
[rank0]: File "/home/otto/.conda/envs/latest/lib/python3.13/site-packages/yaml/representer.py", line 52, in represent_data
[rank0]: node = self.yaml_multi_representers[data_type](self, data)
[rank0]: File "/home/otto/.conda/envs/latest/lib/python3.13/site-packages/yaml/representer.py", line 342, in represent_object
[rank0]: return self.represent_mapping(
[rank0]: ~~~~~~~~~~~~~~~~~~~~~~^
[rank0]: 'tag:yaml.org,2002:python/object:'+function_name, state)
[rank0]: ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[rank0]: File "/home/otto/.conda/envs/latest/lib/python3.13/site-packages/yaml/representer.py", line 118, in represent_mapping
[rank0]: node_value = self.represent_data(item_value)
[rank0]: File "/home/otto/.conda/envs/latest/lib/python3.13/site-packages/yaml/representer.py", line 52, in represent_data
[rank0]: node = self.yaml_multi_representers[data_type](self, data)
[rank0]: File "/home/otto/.conda/envs/latest/lib/python3.13/site-packages/yaml/representer.py", line 330, in represent_object
[rank0]: dictitems = dict(dictitems)
[rank0]: ValueError: dictionary update sequence element #0 has length 1; 2 is required

Based on the discussion in Lightning-AI/pytorch-lightning#9318, it seems that the issue might be related to the save_hyperparameters line.

Could you let me know which Python version you're using? Also, is the requirements.txt file up to date?

Thanks in advance for your help!

Best,
Tim Otto

[harshakokel]

Hello @OttoTim ,

Thank you for the pointer to save_hyperparameter. I was able to replicate the issue and have checked-in a fix for it.

Please let me know if this works.

The python version we are using is 3.9.15 and pytorch-lightning==1.8.6

Regards,
Harsha Kokel.