# Custom Tokenizing?

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fcustom-tokenizing%2F78027%3E%5C
**Category:** 🤗Tokenizers\
**Created:** [March 19, 2024, 4:15am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fcustom-tokenizing%2F78027 "2024-03-19T04:15:31Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![thedarkknight7](https://avatars.discourse-cdn.com/v4/letter/t/e99b99/32.png) [@thedarkknight7](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fthedarkknight7)%5C
**Post date:** [March 19, 2024, 4:15am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fcustom-tokenizing%2F78027%2F1 "2024-03-19T04:15:31Z")

</div>

I have already tokenized my dataset in the desired format for the problem, so I don’t want to tokenize it again. I’m working with nucleotides so I want single, paired and triple sequences. However, I do want to pass this into a BERT model and would like to preprocess the data. I’ve looked at this link here: [Preprocess](https://huggingface.co/docs/transformers/en/preprocessing) (particularly the video). I currently have my tokens stored as a list. Is there anyway I can use that for the remaining steps mentioned in the video (i.e. converting to ids and preparing for model)?

Thanks!