There is an issue with the custom dataset

[zyjhxd]

I encountered an error while training a custom dataset based on s3dis:/root/autodl tpp/SoftGroup/softgroup/evaluation/instance-evaluation. py: 211: Runtime Warning: Mean of empty slice
avg_dict['all_ap'] = np.nanmean(aps[d_inf, :, oAllBut25])
/root/autodl-tmp/SoftGroup/softgroup/evaluation/instance_eval.py:212: RuntimeWarning: Mean of empty slice
avg_dict['all_ap_50%'] = np.nanmean(aps[d_inf, :, o50])
/root/autodl-tmp/SoftGroup/softgroup/evaluation/instance_eval.py:213: RuntimeWarning: Mean of empty slice
avg_dict['all_ap_25%'] = np.nanmean(aps[d_inf, :, o25])
/root/autodl-tmp/SoftGroup/softgroup/evaluation/instance_eval.py:214: RuntimeWarning: Mean of empty slice
avg_dict['all_rc'] = np.nanmean(rcs[d_inf, :, oAllBut25])
/root/autodl-tmp/SoftGroup/softgroup/evaluation/instance_eval.py:215: RuntimeWarning: Mean of empty slice
avg_dict['all_rc_50%'] = np.nanmean(rcs[d_inf, :, o50])
/root/autodl-tmp/SoftGroup/softgroup/evaluation/instance_eval.py:216: RuntimeWarning: Mean of empty slice
avg_dict['all_rc_25%'] = np.nanmean(rcs[d_inf, :, o25])

################################################################
what : AP AP_50% AP_25% AR RC_50% RC_25%
################################################################
ground : nan nan nan nan nan nan
tree : nan nan nan nan nan nan

average : nan nan nan nan nan nan
################################################################

2024-11-15 14:16:41,715 - WARNING - NaN or Inf found in input tensor.
2024-11-15 14:16:41,715 - WARNING - NaN or Inf found in input tensor.
2024-11-15 14:16:41,715 - WARNING - NaN or Inf found in input tensor.
2024-11-15 14:16:41,715 - INFO - AP: nan. AP_50: nan. AP_25: nan, how can I solve it? @thangvubk