bert-arsentd-lev

Arabic version bert model fine tuned on ArSentD-LEV dataset

Data

The model were fine-tuned on ~4000 sentence from twitter multiple dialect and five classes we used 3 out of 5 int the experiment.

Results

class	precision	recall	f1-score	Support
0	0.8211	0.8080	0.8145	125
1	0.7174	0.7857	0.7500	84
2	0.6867	0.6404	0.6628	89
Accuracy			0.7517	298

How to use

You can use these models by installing torch or tensorflow and Huggingface library transformers. And you can use it directly by initializing it like this:

from transformers import AutoModelForSequenceClassification, AutoTokenizer

model_name="mofawzy/bert-arsentd-lev"
model = AutoModelForSequenceClassification.from_pretrained(model_name,num_labels=3)
tokenizer = AutoTokenizer.from_pretrained(model_name)