# What should be processing\_class param value of Seq2SeqTrainer for VisionEncoderDecoderModel Finetuning?

**URL:** <https://discuss.huggingface.co/t/what-should-be-processing-class-param-value-of-seq2seqtrainer-for-visionencoderdecodermodel-finetuning/134101>\
**Category:** 🤗Tokenizers\
**Created:** [January 3, 2025, 12:53pm UTC](https://discuss.huggingface.co/t/what-should-be-processing-class-param-value-of-seq2seqtrainer-for-visionencoderdecodermodel-finetuning/134101 "2025-01-03T12:53:40Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![HarsimarSingh](https://avatars.discourse-cdn.com/v4/letter/h/4bbf92/32.png) [@HarsimarSingh](https://discuss.huggingface.co/u/HarsimarSingh)\
**Post date:** [January 3, 2025, 12:53pm UTC](https://discuss.huggingface.co/t/what-should-be-processing-class-param-value-of-seq2seqtrainer-for-visionencoderdecodermodel-finetuning/134101/1 "2025-01-03T12:53:40Z")

</div>

The image processor and tokenizer along with model is defined as -

image\_processor = AutoImageProcessor.from\_pretrained(“microsoft/swinv2-tiny-patch4-window8-256”)  
tokenizer = XLMRobertaTokenizerFast.from\_pretrained(“FacebookAI/xlm-roberta-base”)  
model = VisionEncoderDecoderModel.from\_encoder\_decoder\_pretrained(  
“microsoft/swinv2-tiny-patch4-window8-256”,  
“FacebookAI/xlm-roberta-base”  
)  
What should be the processing\_class argument for this setup?

Seq2SeqTrainer(  
model=model,  
args=training\_args,  
train\_dataset=train\_dataset,  
eval\_dataset=val\_dataset,  
processing\_class= ? ,  
compute\_metrics=compute\_cer,  
data\_collator=collate\_fn,  
)