metadata

base_model: elnasharomar2/ANER_arabic_keyword_extraction
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: ANER_arabic_keyword_extraction
    results: []

ANER_arabic_keyword_extraction

This model is a fine-tuned version of elnasharomar2/ANER_arabic_keyword_extraction on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.4047
Precision: 0.6061
Recall: 0.6492
F1: 0.6269
Accuracy: 0.9462

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
train_batch_size: 16
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 15
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.0108	1.0	750	0.2997	0.5879	0.6302	0.6083	0.9451
0.0104	2.0	1500	0.2822	0.5699	0.6425	0.6040	0.9428
0.007	3.0	2250	0.3270	0.5965	0.6182	0.6072	0.9446
0.0053	4.0	3000	0.3436	0.5792	0.6439	0.6099	0.9437
0.0038	5.0	3750	0.3373	0.6063	0.6223	0.6142	0.9469
0.0039	6.0	4500	0.3518	0.5961	0.6503	0.6220	0.9462
0.0031	7.0	5250	0.3654	0.5887	0.6488	0.6173	0.9445
0.0029	8.0	6000	0.3985	0.5973	0.6492	0.6222	0.9446
0.0022	9.0	6750	0.3953	0.5927	0.6570	0.6232	0.9456
0.002	10.0	7500	0.3884	0.6145	0.6365	0.6253	0.9474
0.0015	11.0	8250	0.4170	0.5964	0.6566	0.6251	0.9446
0.0015	12.0	9000	0.4421	0.5918	0.6629	0.6253	0.9445
0.0016	13.0	9750	0.4313	0.6078	0.6480	0.6273	0.9465
0.0025	14.0	10500	0.4096	0.6066	0.6432	0.6244	0.9463
0.0023	15.0	11250	0.4047	0.6061	0.6492	0.6269	0.9462

Framework versions

Transformers 4.35.2
Pytorch 2.1.0+cu118
Datasets 2.15.0
Tokenizers 0.15.0