KeyError: 'seq_list'

[elokkk]

Thanks for sharing the code.
When I run: python train.py --trainer mapfe4mp --device_gpu 3
following error occured:
mapfe4mp
[INFO: 132]: Config file: ./config/config_mapfe4mp.yml
[INFO: 492]: Configuration:
[INFO: 493]: {'use_gpu': 1, 'device_gpu': 3, 'dataset_name': 'argoverse_motion_forecasting_dataset', 'dataset': {'path': 'data/datasets/argoverse/motion-forecasting/', 'imgs_folder': 'map_features', 'split': 'train', 'batch_size': 128, 'num_workers': 0, 'start_from_percentage': 0.0, 'split_percentage': 1.0, 'extra_data_train': -1.0, 'hard_mining': -1.0, 'class_balance': -1.0, 'apply_rotation': True, 'shuffle': True, 'data_augmentation': True, 'preprocess_data': False, 'save_data': False}, 'optim_parameters': {'g_learning_rate': 0.001, 'g_weight_decay': 0.001}, 'hyperparameters': {'output_single_agent': True, 'use_rel_disp': True, 'physical_context': 'plausible_centerlines', 'num_modes': 6, 'obs_origin': 20, 'obs_len': 20, 'pred_len': 30, 'distance_threshold': 40, 'num_epochs': 100, 'print_every': 50, 'checkpoint_name': '0', 'checkpoint_val_percentage': 0.0001, 'checkpoint_train_percentage': 0.1, 'loss_type_g': 'nll', 'loss_ade_weight': 1.1, 'loss_fde_weight': 1.4, 'loss_nll_weight': 2.2, 'loss_ade_centerlines_weight': 1, 'loss_smoothL1_weight': 1, 'loss_fa_weight': 1, 'lr_scheduler': True, 'lr_decay_factor': 0.5, 'lr_min': 5e-06, 'lr_epoch_percentage_patience': 0.2, 'g_steps': 1, 'clipping_threshold_g': 1.1, 'save_root_dir': 'save/argoverse', 'exp_name': 'test_save_model', 'output_dir': 'save/argoverse/mapfe4mp/100_percent/test_save_model', 'checkpoint_start_from': None, 'exp_description': None, 'tensorboard_active': True}, 'model': {'path': 'model/models/', 'name': 'mapfe4mp'}, 'base_dir': '/root/autodl-tmp/mapfe4mp'}
[INFO: 500]: Initializing train dataset
files: 208272
Num files to be analyzed: 208272
Loading .npy files as np data structures ...

/root/autodl-tmp/mapfe4mp/model/datasets/argoverse/dataset_utils.py(245)load_processed_files_from_npy()
-> preprocessed_data_dict = dict()
(Pdb) c
Traceback (most recent call last):
File "train.py", line 143, in
model_trainer(config, logger)
File "/root/autodl-tmp/mapfe4mp/model/trainers/trainer_mapfe4mp.py", line 502, in model_trainer
data_train = ArgoverseMotionForecastingDataset(dataset_name=config.dataset_name,
File "/root/autodl-tmp/mapfe4mp/model/datasets/argoverse/dataset.py", line 699, in init
operator.itemgetter(*required_variables_name_list)(preprocess_data_dict)
KeyError: 'seq_list'
How should I solve this problem？