# HF Datasets not working with Language Modeling notebook
**URL:**
**Author:**  [@Trainmaster9977](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FTrainmaster9977)%5C
**Post date:** [May 2, 2021, 2:58am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fhf-datasets-not-working-with-language-modeling-notebook%2F5922%2F1 "2021-05-02T02:58:53Z")
Posted this in the beginners part of the forum, but didnt get any responses so decided to repost here.
I am trying to use this notebook ([https://colab.research.google.com/github/huggingface/notebooks/blob/master/examples/language\_modeling.ipynb](https://colab.research.google.com/github/huggingface/notebooks/blob/master/examples/language_modeling.ipynb)) to finetune and generate text with GPT-Neo using my own custom dataset. I uploaded my own text file to my dataset, but when trying to use it, it just gives me this error
FileNotFoundError Traceback (most recent call last)
/usr/local/lib/python3.7/dist-packages/datasets/load.py in prepare\_module(path, script\_version, download\_config, download\_mode, dataset, force\_local\_path, dynamic\_modules\_path, return\_resolved\_file\_path, \*\*download\_kwargs)
354 try:
→ 355 local\_path = cached\_path(file\_path, download\_config=download\_config)
356 except FileNotFoundError:
4 frames
FileNotFoundError: Couldn’t find file at htts://huggingface.co/datasets/Trainmaster9977/zbakuman/resolve/main/zbakuman.py
During handling of the above exception, another exception occurred:
FileNotFoundError Traceback (most recent call last)
/usr/local/lib/python3.7/dist-packages/datasets/load.py in prepare\_module(path, script\_version, download\_config, download\_mode, dataset, force\_local\_path, dynamic\_modules\_path, return\_resolved\_file\_path, \*\*download\_kwargs)
357 raise FileNotFoundError(
358 “Couldn’t find file locally at {}, or remotely at {}. Please provide a valid {} name”.format(
→ 359 combined\_path, file\_path, “dataset” if dataset else “metric”
360 )
361 )
FileNotFoundError: Couldn’t find file locally at Trainmaster9977/zbakuman/zbakuman.py, or remotely at htts://huggingface.co/datasets/Trainmaster9977/zbakuman/resolve/main/zbakuman.py. Please provide a valid dataset name
(Slightly modified beginning of url due to new users only able to put 2 links in posts)
I tried to use this to load it
from datasets import load\_dataset
dataset = load\_dataset(“Trainmaster9977/zbakuman”)
---
**Author:**  [@lewtun](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Flewtun)%5C
**Post date:** [May 2, 2021, 10:07am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fhf-datasets-not-working-with-language-modeling-notebook%2F5922%2F2 "2021-05-02T10:07:49Z")
hey @Trainmaster9977, as described in the [docs](https://huggingface.co/docs/datasets/loading_datasets.html#from-local-files) you need to provide an argument to `load_dataset` that indicates the file format (csv, json, etc).
ps. in future please don’t create duplicate posts (either edit the original one or delete it if necessary)
---
**Author:**  [@Trainmaster9977](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FTrainmaster9977)%5C
**Post date:** [May 2, 2021, 5:52pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fhf-datasets-not-working-with-language-modeling-notebook%2F5922%2F3 "2021-05-02T17:52:45Z")
ok, thank you, and noted.