POEMS-CAMELBERT-CA-RUN4-25

This model is a fine-tuned version of CAMeL-Lab/bert-base-arabic-camelbert-ca on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.7725
• Accuracy: 0.5731
• F1: 0.5731
• Precision: 0.5731
• Recall: 0.5731

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: 5e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
1.3427	1.0	472	1.2734	0.4056	0.4056	0.4056	0.4056
1.1941	2.0	944	1.1317	0.5129	0.5129	0.5129	0.5129
1.1178	3.0	1416	1.1461	0.5142	0.5142	0.5142	0.5142
1.0569	4.0	1888	1.0592	0.5412	0.5412	0.5412	0.5412
0.9925	5.0	2360	1.1219	0.5426	0.5426	0.5426	0.5426
0.9375	6.0	2832	1.0840	0.5740	0.5740	0.5740	0.5740
0.8771	7.0	3304	1.1091	0.5816	0.5816	0.5816	0.5816
0.8262	8.0	3776	1.1221	0.5851	0.5851	0.5851	0.5851
0.7871	9.0	4248	1.1499	0.5745	0.5745	0.5745	0.5745
0.7252	10.0	4720	1.3011	0.5621	0.5621	0.5621	0.5621
0.6919	11.0	5192	1.3272	0.5802	0.5802	0.5802	0.5802
0.6427	12.0	5664	1.3928	0.5683	0.5683	0.5683	0.5683
0.6008	13.0	6136	1.4789	0.5590	0.5590	0.5590	0.5590
0.5576	14.0	6608	1.4850	0.5638	0.5638	0.5638	0.5638
0.5267	15.0	7080	1.5124	0.5762	0.5762	0.5762	0.5762
0.4823	16.0	7552	1.3870	0.5683	0.5683	0.5683	0.5683
0.4564	17.0	8024	1.5277	0.5785	0.5785	0.5785	0.5785
0.4217	18.0	8496	1.5805	0.5723	0.5723	0.5723	0.5723
0.3891	19.0	8968	1.5173	0.5709	0.5709	0.5709	0.5709
0.3705	20.0	9440	1.6484	0.5807	0.5807	0.5807	0.5807
0.3419	21.0	9912	1.6999	0.5816	0.5816	0.5816	0.5816
0.321	22.0	10384	1.7024	0.5745	0.5745	0.5745	0.5745
0.3121	23.0	10856	1.7545	0.5709	0.5709	0.5709	0.5709
0.2964	24.0	11328	1.7355	0.5767	0.5767	0.5767	0.5767
0.285	25.0	11800	1.7725	0.5731	0.5731	0.5731	0.5731

Framework versions

• Transformers 4.38.2
• Pytorch 2.1.0+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2