POEMS-CAMELBERT-CA-RUN4-20-fullData

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: 3.1172
• Accuracy: 0.6210
• F1: 0.6210
• Precision: 0.6210
• Recall: 0.6210

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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
1.1804	1.0	568	1.1103	0.5066	0.5066	0.5066	0.5066
0.9771	2.0	1136	0.9937	0.5847	0.5847	0.5847	0.5847
0.8057	3.0	1704	1.0751	0.5882	0.5882	0.5882	0.5882
0.6404	4.0	2272	1.1029	0.6011	0.6011	0.6011	0.6011
0.4956	5.0	2840	1.1222	0.6064	0.6064	0.6064	0.6064
0.3742	6.0	3408	1.2714	0.6077	0.6077	0.6077	0.6077
0.2881	7.0	3976	1.5337	0.5931	0.5931	0.5931	0.5931
0.2153	8.0	4544	1.6150	0.5984	0.5984	0.5984	0.5984
0.1663	9.0	5112	1.7246	0.6037	0.6037	0.6037	0.6037
0.1266	10.0	5680	2.0767	0.5984	0.5984	0.5984	0.5984
0.1064	11.0	6248	2.1690	0.6161	0.6161	0.6161	0.6161
0.0895	12.0	6816	2.4732	0.6068	0.6068	0.6068	0.6068
0.0794	13.0	7384	2.4153	0.6095	0.6095	0.6095	0.6095
0.0555	14.0	7952	2.8754	0.6037	0.6037	0.6037	0.6037
0.0502	15.0	8520	2.8673	0.6121	0.6121	0.6121	0.6121
0.0383	16.0	9088	2.9805	0.6139	0.6139	0.6139	0.6139
0.0329	17.0	9656	3.0402	0.6188	0.6188	0.6188	0.6188
0.0237	18.0	10224	3.0225	0.6263	0.6263	0.6263	0.6263
0.0173	19.0	10792	3.0629	0.6206	0.6206	0.6206	0.6206
0.0169	20.0	11360	3.1172	0.6210	0.6210	0.6210	0.6210

Framework versions

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