# Different BERT results

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fdifferent-bert-results%2F18258%3E%5C
**Category:** Beginners\
**Created:** [May 23, 2022, 10:38am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fdifferent-bert-results%2F18258 "2022-05-23T10:38:06Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![mox](https://avatars.discourse-cdn.com/v4/letter/m/f04885/32.png) [@mox](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fmox)%5C
**Post date:** [May 23, 2022, 10:38am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fdifferent-bert-results%2F18258%2F1 "2022-05-23T10:38:06Z")

</div>

Hello,

i have trained different BERT models to evaluate which of these models perform best with my dataset.  
At the moment I got the problem that I always get different results when running the model again (e.g. run a new colab session).  
For a reproducible evaluation I want to get always the same results (e.g. F1) but I don’t know how to do this.  
I have ensured that my train/test/validation split is always the same.  
Here is an example notebook: [Google Colab](https://colab.research.google.com/drive/1OLaKl5RmO9GKIpfewtg5sIxCJpKJE_4F?usp=sharing)

Thanks in advance!

---

<div class="post-metadata">

**Author:** ![torbotis](https://avatars.discourse-cdn.com/v4/letter/t/e480ec/32.png) [@torbotis](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Ftorbotis)%5C
**Post date:** [May 25, 2022, 10:36am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fdifferent-bert-results%2F18258%2F2 "2022-05-25T10:36:04Z")

</div>

used seed to make your results reproducible.

```auto
random.seed(seed)
    np.random.seed(seed)
    if is_torch_available():
        tt.manual_seed(seed)
        tt.cuda.manual_seed_all(seed)

```