Modelpublic

MM2157/AraBERT_token_classification__AraEval24_aug_rand_concat

sourceHugging Faceupdated 2y agoView on Hugging Face

0likes18downloads

Model Card

AraBERTtokenclassification_AraEval24augrandconcat

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: 1.4270
—Precision: 0.0167
—Recall: 0.0238
—F1: 0.0196
—Accuracy: 0.6741

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
1.9197	1.0	938	1.1953	0.0093	0.0092	0.0092	0.6946
0.8179	2.0	1876	1.1406	0.0018	0.0019	0.0018	0.6789
0.6136	3.0	2814	1.1013	0.0125	0.0136	0.0131	0.7152
0.4945	4.0	3752	1.1583	0.0097	0.0110	0.0103	0.6996
0.4105	5.0	4690	1.2239	0.0140	0.0182	0.0158	0.6816
0.3536	6.0	5628	1.3073	0.0155	0.0214	0.0180	0.6658
0.3097	7.0	6566	1.3764	0.0147	0.0208	0.0172	0.6574
0.2729	8.0	7504	1.3447	0.0141	0.0192	0.0162	0.6810
0.2525	9.0	8442	1.4392	0.0160	0.0234	0.0190	0.6629
0.2393	10.0	9380	1.4270	0.0167	0.0238	0.0196	0.6741

Framework versions

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