RecBole: ValueError: `training_neg_sample_num` should be 0 if inter_feat have label_field [label].

I am getting the following error for LightGCN

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-14-0b2a29229228> in <module>
     47 
     48 # dataset splitting
---> 49 train_data, valid_data, test_data = data_preparation(config, dataset)
     50 
     51 # model loading and initialization

c:\users\laxmi\downloads\recbole\recbole\data\utils.py in data_preparation(config, dataset, save)
    101     if config['training_neg_sample_num']:
    102         if dataset.label_field in dataset.inter_feat:
--> 103             raise ValueError(f'`training_neg_sample_num` should be 0 '
    104                              f'if inter_feat have label_field [{dataset.label_field}].')
    105         train_distribution = config['training_neg_sample_distribution'] or 'uniform'

ValueError: `training_neg_sample_num` should be 0 if inter_feat have label_field [label].

If I set training_neg_sample_num=0 then I get another error saying

c:\users\laxmi\downloads\recbole\recbole\model\general_recommender\lightgcn.py in calculate_loss(self, interaction)
    136         user = interaction[self.USER_ID]
    137         pos_item = interaction[self.ITEM_ID]
--> 138         neg_item = interaction[self.NEG_ITEM_ID]
    139 
    140         user_all_embeddings, item_all_embeddings = self.forward()

c:\users\laxmi\downloads\recbole\recbole\data\interaction.py in __getitem__(self, index)
    103     def __getitem__(self, index):
    104         if isinstance(index, str):
--> 105             return self.interaction[index]
    106         else:
    107             ret = {}

KeyError: 'neg_pid'

Any idea, how to remove this error?