# Seq2SeqTrainer with num\_beams and generation\_config

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fseq2seqtrainer-with-num-beams-and-generation-config%2F41549%3E%5C
**Category:** 🤗Transformers\
**Created:** [May 31, 2023, 2:55am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fseq2seqtrainer-with-num-beams-and-generation-config%2F41549 "2023-05-31T02:55:53Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![anon99315630](https://avatars.discourse-cdn.com/v4/letter/a/dec6dc/32.png) [@anon99315630](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fanon99315630)%5C
**Post date:** [May 31, 2023, 2:55am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fseq2seqtrainer-with-num-beams-and-generation-config%2F41549%2F1 "2023-05-31T02:55:53Z")

</div>

I am using a `Seq2SeqTrainer` for evaluating a translation task. I found that the `Seq2SeqTrainingArguments` accept the `generation_num_beams` argument and several other arguments for generation. Meanwhile, it has the `generation_config` argument, which can config the `num_beams` too. So I wonder what are the difference between these arguments, will one setting in `Seq2SeqTrainingArguments` override the counterpart setting in `generation_config`?