ArabicNewSplits6_FineTuningAraBERT_run2_AugV5_k1_task2_organization

This model is a fine-tuned version of aubmindlab/bert-base-arabertv02 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.6518
• Qwk: 0.4745
• Mse: 0.6518
• Rmse: 0.8073

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Qwk	Mse	Rmse
No log	0.2857	2	4.0222	-0.0279	4.0222	2.0055
No log	0.5714	4	2.2102	0.0914	2.2102	1.4867
No log	0.8571	6	1.1316	0.0773	1.1316	1.0638
No log	1.1429	8	0.7754	0.0358	0.7754	0.8805
No log	1.4286	10	0.6912	0.2686	0.6912	0.8314
No log	1.7143	12	0.8298	0.1677	0.8298	0.9109
No log	2.0	14	1.0315	0.1883	1.0315	1.0156
No log	2.2857	16	0.9664	0.1956	0.9664	0.9830
No log	2.5714	18	0.7434	0.2228	0.7434	0.8622
No log	2.8571	20	0.6926	0.1862	0.6926	0.8322
No log	3.1429	22	0.7192	0.1011	0.7192	0.8480
No log	3.4286	24	0.6465	0.2247	0.6465	0.8040
No log	3.7143	26	0.6199	0.2326	0.6199	0.7873
No log	4.0	28	0.6086	0.2196	0.6086	0.7801
No log	4.2857	30	0.6035	0.3198	0.6035	0.7769
No log	4.5714	32	0.5989	0.3289	0.5989	0.7739
No log	4.8571	34	0.5946	0.3762	0.5946	0.7711
No log	5.1429	36	0.6082	0.3744	0.6082	0.7799
No log	5.4286	38	0.6665	0.3761	0.6665	0.8164
No log	5.7143	40	0.6983	0.3933	0.6983	0.8356
No log	6.0	42	0.6270	0.4017	0.6270	0.7919
No log	6.2857	44	0.6177	0.4374	0.6177	0.7859
No log	6.5714	46	0.6346	0.4438	0.6346	0.7966
No log	6.8571	48	0.6237	0.4471	0.6237	0.7898
No log	7.1429	50	0.6222	0.4496	0.6222	0.7888
No log	7.4286	52	0.6200	0.4659	0.6200	0.7874
No log	7.7143	54	0.6171	0.4686	0.6171	0.7856
No log	8.0	56	0.6212	0.4765	0.6212	0.7881
No log	8.2857	58	0.6289	0.4804	0.6289	0.7930
No log	8.5714	60	0.6405	0.4663	0.6405	0.8003
No log	8.8571	62	0.6485	0.4663	0.6485	0.8053
No log	9.1429	64	0.6503	0.4449	0.6503	0.8064
No log	9.4286	66	0.6502	0.4738	0.6502	0.8064
No log	9.7143	68	0.6516	0.4745	0.6516	0.8072
No log	10.0	70	0.6518	0.4745	0.6518	0.8073

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0+cu118
• Datasets 2.21.0
• Tokenizers 0.19.1