# Inconsistent evaluation result

**URL:** <https://discuss.huggingface.co/t/inconsistent-evaluation-result/113480>\
**Category:** Beginners\
**Created:** [October 23, 2024, 1:07am UTC](https://discuss.huggingface.co/t/inconsistent-evaluation-result/113480 "2024-10-23T01:07:32Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![why341](https://avatars.discourse-cdn.com/v4/letter/w/f08c70/32.png) [@why341](https://discuss.huggingface.co/u/why341)\
**Post date:** [October 23, 2024, 1:07am UTC](https://discuss.huggingface.co/t/inconsistent-evaluation-result/113480/1 "2024-10-23T01:07:32Z")

</div>

> **[AppliedML123/whisper-large-v3\_ADReSSo · Hugging Face](https://huggingface.co/AppliedML123/whisper-large-v3_ADReSSo)**
>
> We’re on a journey to advance and democratize artificial intelligence through open source and open science.

 ![image](https://us1.discourse-cdn.com/hellohellohello/original/3X/0/a/0a652f62a2080b156737b07a8387918bf412b185.png)

Hi, guys. I’m learning how to train a transformer here. My accuracy initially is 0.78, then it jumps to 0.85, and 0.88 when I run  
eval\_result = trainer.evaluate() for a few times. Can anyone help me to have a look and tell me what’s the problem here? And if so, how can I solve this?