# Issue in using trainer class for Finetuning GPT-2 **URL:** \ **Category:** Models\ **Created:** [November 21, 2020, 12:39pm UTC](https://discuss.huggingface.co/t/issue-in-using-trainer-class-for-finetuning-gpt-2/2155 "2020-11-21T12:39:10Z")\ **Posts on this page:** 2\ **Page:** 1
**Author:** ![Parth](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/parth/32/401_2.png) [@Parth](https://discuss.huggingface.co/u/Parth)\ **Post date:** [November 21, 2020, 12:39pm UTC](https://discuss.huggingface.co/t/issue-in-using-trainer-class-for-finetuning-gpt-2/2155/1 "2020-11-21T12:39:11Z")
While using trainer class to finetune gpt-2 on Hindi dataset. It’s outputting the following error While using trainer class to finetune gpt-2 on Hindi dataset. It’s outputting the following error ```auto TypeError Traceback (most recent call last) in () ----> 1 trainer.train() 5 frames /usr/local/lib/python3.6/dist-packages/transformers/data/datasets/language_modeling.py in __getitem__ (self, i) 99 100 def __getitem__ (self, i) -> torch.Tensor: --> 101 return torch.tensor(self.examples[i], dtype=torch.long) 102 103 TypeError: an integer is required (got type NoneType) ``` Here is the link: [https://colab.research.google.com/drive/1um5UeY9hasmjPNcR1WkBe2uDDFhLUBrX?usp=sharing](https://colab.research.google.com/drive/1um5UeY9hasmjPNcR1WkBe2uDDFhLUBrX?usp=sharing) ---
**Author:** ![Parth](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/parth/32/401_2.png) [@Parth](https://discuss.huggingface.co/u/Parth)\ **Post date:** [November 23, 2020, 11:37am UTC](https://discuss.huggingface.co/t/issue-in-using-trainer-class-for-finetuning-gpt-2/2155/2 "2020-11-23T11:37:35Z")
@valhalla, Hey can you suggest something about what might be the problem. I am using trainer class on Hindi trained GPT-2. I have also shared the Colab.