I have followed all the steps mentioned in the repo and the training went smooth enough until this point. Most probably it has a bug in rasat/seq2seq/utils/spider.py. Here is the error description:
ValueError: setting an array element with a sequence. The requested array has an inhomogeneous shape after 1 dimensions. The detected shape was (1034,) + inhomogeneous part.
After examining the spider.py, I have noticed the comment "# Replace -100 in the labels as we can't decode them.". Which value the label_ids should be replaced?
I have followed all the steps mentioned in the repo and the training went smooth enough until this point. Most probably it has a bug in rasat/seq2seq/utils/spider.py. Here is the error description:
Can you please help me to fix the issue?