# Electra NER on Conll03-english
**URL:**
**Author:**  [@sergulaydore](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fsergulaydore)%5C
**Post date:** [September 8, 2020, 3:14am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Felectra-ner-on-conll03-english%2F1032%2F1 "2020-09-08T03:14:13Z")
Hi all,
I am new to NLP. I would like to understand how model dbmdz/electra-large-discriminator-finetuned-conll03-english was fine tuned on Conll03-english dataset. Is there any available code that I can take a look? I understand it was trained by [@stefan-it](https://github.com/stefan-it).
Thank you,
Sergul
---
**Author:**  [@valhalla](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fvalhalla)%5C
**Post date:** [September 8, 2020, 6:09am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Felectra-ner-on-conll03-english%2F1032%2F2 "2020-09-08T06:09:16Z")
Hi @sergulaydore you can find the ner training code [here](https://github.com/huggingface/transformers/tree/master/examples/token-classification)
---
**Author:**  [@sergulaydore](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fsergulaydore)%5C
**Post date:** [September 8, 2020, 3:50pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Felectra-ner-on-conll03-english%2F1032%2F3 "2020-09-08T15:50:45Z")
Thank you @valhalla but I am specifically asking for the electra model. The code you refer to uses bert-base-multilingual-cased as far as I understand.
---
**Author:**  [@Elidor00](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FElidor00)%5C
**Post date:** [September 8, 2020, 5:08pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Felectra-ner-on-conll03-english%2F1032%2F4 "2020-09-08T17:08:24Z")
@sergulaydore I think it is enough to change the name of the pretrained model (`--model_name_or_path`) in `.sh` script you use to run `run_ner.py`.
Check how to [Run the Pytorch version](https://github.com/huggingface/transformers/blob/master/examples/token-classification/README.md#run-the-pytorch-version).
---
**Author:**  [@sergulaydore](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fsergulaydore)%5C
**Post date:** [September 8, 2020, 5:26pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Felectra-ner-on-conll03-english%2F1032%2F5 "2020-09-08T17:26:11Z")
But I don’t know what was the pre-trained model used for this dbmdz/electra-large-discriminator-finetuned-conll03-english .
---
**Author:**  [@valhalla](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fvalhalla)%5C
**Post date:** [September 9, 2020, 12:20pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Felectra-ner-on-conll03-english%2F1032%2F6 "2020-09-09T12:20:13Z")
@Elidor00 is right! You’ll just need to change the model name argument
as you can guess from the name `dbmdz/electra-large-discriminator-finetuned-conll03-english` uses `google/electra-large-discriminator` pre-trained model.
Most of the models follow this naming format
`pretrainedmodelname-finetuned-task-or-dataset-name`
---
**Author:**  [@sergulaydore](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fsergulaydore)%5C
**Post date:** [September 9, 2020, 1:45pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Felectra-ner-on-conll03-english%2F1032%2F7 "2020-09-09T13:45:52Z")
Oh ok, thank you!