TypeError when train the vanilla ManiGaussian

[DingChunQ]

When I run the vanilla ManiGaussian using: "bash scripts/train_and_eval_w_geo_sem_dyna.sh ManiGaussian_BC 0,1 12345 ${exp_name}"

The following error occurs:
L_rot: 16.769 x 1.000 | L_grip: 0.535 x 1.000 | L_col: 0.328 x 1.000 | L_rgb: 0.041 x 0.010 | L_embed: 0.780 x 0.0001 | L_dyna: 0.094 x 0.0000 | L_reg: 0.000 x 0.0000 | psnr: 13.879
0%|▎ | 285/100010 [16:30<96:17:34, 3.48s/it]
Error executing job with overrides: ['method=ManiGaussian_BC', 'rlbench.task_name=vanilla_ManiGaussian_0', 'rlbench.demo_path=/home/aida/wangyc/ManiGaussian/data/train_data', 'replay.path=/home/aida/wangyc/ManiGaussian/replay/vanilla_ManiGaussian_0', 'framework.start_seed=0', 'framework.use_wandb=True', 'method.use_wandb=True', 'framework.wandb_group=vanilla_ManiGaussian_0', 'framework.wandb_name=vanilla_ManiGaussian_0', 'ddp.num_devices=2', 'replay.batch_size=1', 'ddp.master_port=12345', 'rlbench.tasks=[close_jar,open_drawer,sweep_to_dustpan_of_size,meat_off_grill,turn_tap,slide_block_to_color_target,put_item_in_drawer,reach_and_drag,push_buttons,stack_blocks]', 'rlbench.demos=20', 'method.neural_renderer.render_freq=2000', 'method.neural_renderer.lambda_embed=0.01', 'method.neural_renderer.lambda_dyna=0.1', 'method.neural_renderer.lambda_reg=0.0', 'method.neural_renderer.foundation_model_name=diffusion', 'method.neural_renderer.use_dynamic_field=True']
Traceback (most recent call last):
File "/home/aida/wangyc/ManiGaussian/train.py", line 96, in main
run_seed_fn.run_seed(
File "/home/aida/wangyc/ManiGaussian/run_seed_fn.py", line 147, in run_seed
train_runner.start()
File "/home/aida/wangyc/ManiGaussian/third_party/YARR/yarr/runners/offline_train_runner.py", line 200, in start
batch = self.preprocess_data(data_iter)
File "/home/aida/wangyc/ManiGaussian/third_party/YARR/yarr/runners/offline_train_runner.py", line 121, in preprocess_data
sampled_batch = next(data_iter) # may raise StopIteration
File "/media/aida/data2/anaconda/envs/manigaussian/lib/python3.9/site-packages/lightning/fabric/wrappers.py", line 296, in iter
for item in self._dataloader:
File "/media/aida/data2/anaconda/envs/manigaussian/lib/python3.9/site-packages/torch/utils/data/dataloader.py", line 634, in next
data = self._next_data()
File "/media/aida/data2/anaconda/envs/manigaussian/lib/python3.9/site-packages/torch/utils/data/dataloader.py", line 678, in _next_data
data = self._dataset_fetcher.fetch(index) # may raise StopIteration
File "/media/aida/data2/anaconda/envs/manigaussian/lib/python3.9/site-packages/torch/utils/data/_utils/fetch.py", line 41, in fetch
data = next(self.dataset_iter)
File "/home/aida/wangyc/ManiGaussian/third_party/YARR/yarr/replay_buffer/wrappers/pytorch_replay_buffer.py", line 17, in _generator
yield self._replay_buffer.sample_transition_batch(pack_in_dict=True)
File "/home/aida/wangyc/ManiGaussian/third_party/YARR/yarr/replay_buffer/uniform_replay_buffer.py", line 722, in sample_transition_batch
store = self._get_from_disk(
File "/home/aida/wangyc/ManiGaussian/third_party/YARR/yarr/replay_buffer/uniform_replay_buffer.py", line 391, in _get_from_disk
d = pickle.load(f)
TypeError: setstate() argument 1 must be sequence of length 4, not 2

Set the environment variable HYDRA_FULL_ERROR=1 for a complete stack trace.
wandb: Waiting for W&B process to finish... (failed 1). Press Control-C to abort syncing.
0%|▎ | 285/100010 [16:50<93:58:23, 3.39s/it]wandb:

Any suggestions for this? Thanks!