Instructions to use HooshvareLab/bert-base-parsbert-uncased with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use HooshvareLab/bert-base-parsbert-uncased with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("fill-mask", model="HooshvareLab/bert-base-parsbert-uncased")# Load model directly from transformers import AutoTokenizer, AutoModelForMaskedLM tokenizer = AutoTokenizer.from_pretrained("HooshvareLab/bert-base-parsbert-uncased") model = AutoModelForMaskedLM.from_pretrained("HooshvareLab/bert-base-parsbert-uncased", device_map="auto") - Inference
- Notebooks
- Google Colab
- Kaggle
Multiple Targets Problem
#2
by amirtavakkoli13 - opened
Hello, I hope you are doing great. I appreciate you providing and sharing this dataset.
There is a catch with some of the rows which contain more than one corresponding Persian target.
First, it complicates the code when used for language modeling(which must be taken care of in these edge cases), and second,
it is not well-stratified.
There is only a single multiple-target row in the training dataset but more than 6000 in the testing dataset.
Thanks for reading this.