Open CaffreyR opened 2 years ago
Hi Rabeeh, in your code, you calculate the total_params_ratio by using this code.
total_params_ratio
total_params_ratio = ((total_params-t5_base_params)*8+t5_base_params)/t5_base_params
https://github.com/rabeehk/compacter/blob/main/seq2seq/utils/utils.py#L213
Why do your multiply 8 instead of just divide? Many thanks
Hi Rabeeh, in your code, you calculate the
total_params_ratioby using this code.total_params_ratio = ((total_params-t5_base_params)*8+t5_base_params)/t5_base_paramshttps://github.com/rabeehk/compacter/blob/main/seq2seq/utils/utils.py#L213
Why do your multiply 8 instead of just divide? Many thanks