Instructions to use HuggingFaceFW/fineweb-edu-classifier with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use HuggingFaceFW/fineweb-edu-classifier with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="HuggingFaceFW/fineweb-edu-classifier")# Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("HuggingFaceFW/fineweb-edu-classifier") model = AutoModelForSequenceClassification.from_pretrained("HuggingFaceFW/fineweb-edu-classifier") - Inference
- Notebooks
- Google Colab
- Kaggle
classification or regression? how to choose.
#7
by DaleMeng - opened
Hello,
According to the code in https://github.com/huggingface/cosmopedia/blob/main/classification/train_edu_bert.py, I find that the model is trained based on a regression model instead of a 6-labels classification model. I wonder if there is any deep insight to do it in a regression way? or have you compared it to the classification way?
I used classification and it resulted in a better score for my use case. Not sure why regression was used?