Modelpublic

MM2157/AraBERT_token_classification__AraEval24_merged_rassd_aratweets

sourceHugging Faceupdated 2y agoView on Hugging Face

0likes24downloads

Model Card

AraBERTtokenclassification_AraEval24mergedrassdaratweets

This model is a fine-tuned version of aubmindlab/bert-base-arabert on the None dataset. It achieves the following results on the evaluation set:

—Loss: 0.8783
—Precision: 0.0736
—Recall: 0.0243
—F1: 0.0365
—Accuracy: 0.8564

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

—learning_rate: 2e-05
—trainbatchsize: 8
—evalbatchsize: 8
—seed: 42
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.7095	1.0	3105	0.8134	1.0	0.0001	0.0002	0.8633
0.6521	2.0	6210	0.7728	0.1149	0.0021	0.0041	0.8631
0.5857	3.0	9315	0.7770	0.0383	0.0009	0.0017	0.8632
0.5233	4.0	12420	0.7929	0.0896	0.0100	0.0180	0.8624
0.5096	5.0	15525	0.7911	0.0716	0.0108	0.0187	0.8617
0.4685	6.0	18630	0.8200	0.0906	0.0144	0.0248	0.8618
0.4393	7.0	21735	0.8399	0.0939	0.0160	0.0273	0.8618
0.4204	8.0	24840	0.8361	0.0862	0.0230	0.0363	0.8590
0.3872	9.0	27945	0.8706	0.0782	0.0251	0.0380	0.8567
0.3569	10.0	31050	0.8783	0.0736	0.0243	0.0365	0.8564

Framework versions

—Transformers 4.30.2
—Pytorch 1.12.1
—Datasets 2.13.2
—Tokenizers 0.13.3