how to generate attention mask seperately?

[Liujingxiu23]

I want to use my own input_embedings(or my own tokenizer system) instead of input_ids, but I do not know how to generate encoder_attention_mask and decoder_attention_mask bymyself, is there any code?

[vishugupta96]

facing the same problem not much info is provided on the inner working of this model making customization difficult.

[DamithDR]

You may refer his other repository : https://github.com/Shivanandroy/T5-Finetuning-PyTorch