Getting Invalid Argument error while training on SPIDER dataset (trying on A100)

[shabbie]

I am following the instructions in readme file to replicate the results on SPIDER data. While the training is going on I am repeatedly getting this error : "tensorflow.python.framework.errors_impl.InvalidArgumentError: Received a label value of 6 which is outside the valid range of [0, 6). Label values: 6" which is orginating from line number 270 inside sqlgen.py file. I believe the labels for the classifier has been generated by parsing the SQL query, is there any issue with the label generation process? or am I missing on something?