# Evaluating pretrained model

**URL:** <https://discuss.huggingface.co/t/evaluating-pretrained-model/8704>\
**Category:** Beginners\
**Created:** [July 26, 2021, 6:25pm UTC](https://discuss.huggingface.co/t/evaluating-pretrained-model/8704 "2021-07-26T18:25:45Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![ariafariakh](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/ariafariakh/32/1952_2.png) [@ariafariakh](https://discuss.huggingface.co/u/ariafariakh)\
**Post date:** [July 26, 2021, 6:25pm UTC](https://discuss.huggingface.co/t/evaluating-pretrained-model/8704/1 "2021-07-26T18:25:45Z")

</div>

hello,  
I see that we can fine tune a model and then evaluate it using certain metrics in chapter3. I am curious to know if it is possible to evaluate the accuracy of the original pretrained model, without the finetuning? That way I can see how much fine-tuning actually helped the model

please let me know

Thank you