ArabicNewSplits4_WithDuplicationsForScore5_FineTuningAraBERT_run2_AugV5_k2_task5_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: 0.5430
• Qwk: 0.7829
• Mse: 0.5430
• Rmse: 0.7369

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	0.2	2	2.3856	0.0137	2.3856	1.5445
No log	0.4	4	1.5110	0.2115	1.5110	1.2292
No log	0.6	6	1.4728	0.1451	1.4728	1.2136
No log	0.8	8	1.5747	0.1220	1.5747	1.2549
No log	1.0	10	1.6031	0.3091	1.6031	1.2661
No log	1.2	12	1.6592	0.3283	1.6592	1.2881
No log	1.4	14	1.3748	0.2121	1.3748	1.1725
No log	1.6	16	1.3393	0.2590	1.3393	1.1573
No log	1.8	18	1.2636	0.3111	1.2636	1.1241
No log	2.0	20	1.2281	0.4374	1.2281	1.1082
No log	2.2	22	1.4192	0.4446	1.4192	1.1913
No log	2.4	24	1.3988	0.5011	1.3988	1.1827
No log	2.6	26	0.9919	0.5418	0.9919	0.9960
No log	2.8	28	0.7729	0.6026	0.7729	0.8792
No log	3.0	30	0.7551	0.6072	0.7551	0.8690
No log	3.2	32	0.7052	0.6170	0.7052	0.8397
No log	3.4	34	0.7365	0.6997	0.7365	0.8582
No log	3.6	36	0.9602	0.6526	0.9602	0.9799
No log	3.8	38	0.9188	0.6819	0.9188	0.9585
No log	4.0	40	0.6862	0.7348	0.6862	0.8284
No log	4.2	42	0.6324	0.7386	0.6324	0.7952
No log	4.4	44	0.6036	0.7386	0.6036	0.7769
No log	4.6	46	0.6135	0.7499	0.6135	0.7833
No log	4.8	48	0.6624	0.7267	0.6624	0.8139
No log	5.0	50	0.6149	0.7638	0.6149	0.7842
No log	5.2	52	0.5465	0.7751	0.5465	0.7392
No log	5.4	54	0.5683	0.7279	0.5683	0.7539
No log	5.6	56	0.5587	0.7451	0.5587	0.7475
No log	5.8	58	0.5594	0.7715	0.5594	0.7479
No log	6.0	60	0.7257	0.7414	0.7257	0.8519
No log	6.2	62	0.8401	0.7183	0.8401	0.9166
No log	6.4	64	0.7631	0.7187	0.7631	0.8735
No log	6.6	66	0.6196	0.7384	0.6196	0.7871
No log	6.8	68	0.5511	0.7720	0.5511	0.7423
No log	7.0	70	0.5501	0.7497	0.5501	0.7417
No log	7.2	72	0.5582	0.7492	0.5582	0.7471
No log	7.4	74	0.5552	0.7497	0.5552	0.7451
No log	7.6	76	0.5545	0.7680	0.5545	0.7447
No log	7.8	78	0.5831	0.7654	0.5831	0.7636
No log	8.0	80	0.6278	0.7525	0.6278	0.7923
No log	8.2	82	0.6377	0.7522	0.6377	0.7986
No log	8.4	84	0.6063	0.7553	0.6063	0.7786
No log	8.6	86	0.5607	0.7666	0.5607	0.7488
No log	8.8	88	0.5484	0.7746	0.5484	0.7406
No log	9.0	90	0.5442	0.7720	0.5442	0.7377
No log	9.2	92	0.5456	0.7795	0.5456	0.7387
No log	9.4	94	0.5441	0.7886	0.5441	0.7376
No log	9.6	96	0.5438	0.7886	0.5438	0.7374
No log	9.8	98	0.5428	0.7829	0.5428	0.7368
No log	10.0	100	0.5430	0.7829	0.5430	0.7369

Framework versions

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